(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The mammals of the Caucasus; a history of the evolution of the fauna. (Mlekopitayushchie Kavkaza; istoriya formirovaniya fauny)"



*C''' икЫ^- ^^i^^-—- . 



N.K.Vereshchagin 



THE MAMMALS OF THE CAUCASUS 

A History of the Evolution of the Fauna 



TRANSLATED FROM RUSSIAN 

Published for the Smithsonian Institution, U.S.A. 

and the National Science Foundation, Washington, D.C. 

by the Israel Program for Scientific Translations 



AKADEMIYA NAUK SSSR . ZOOLOGICHESKII INSTITUT 

AKADEMIYA NAUK AZERBAIDZHANSKOI SSR 

Academy of Sciences of the USSR . Zoological Institute. Academy 
of Sciences of the Azerbaidzhan SSR 



N.K. VERESHCHAGIN 

THE MAMMALS 
OF THE CAUCASUS 

A History of the Evolution of the Fauna 

(Mlekopitayushchie Kavkaza )(Istoriya formirovaniya fauny) 



Chief Editor- E,N, Pavlovskii 
Editor of this volume. A. A. Strelkov 



Izdatel'stvo Akademii Nauk SSSR 
Moskva-Leningrad 1959 



Translated from Russian 



Israel Program for Scientific Translations 
Jerusalem 1967 



TT66-51143 

Published Pursuant to an Agreement with 
THE SMITHSONIAN INSTITUTION, U.S.A. 
and 
THE NATIONAL SCIENCE FOUNDATION, WASHINGTON, D. С 



Copyright © 1967 

Israel Program for Scientific Translations Ltd. 

IPST Cat. No. 1704 



Translated by Dr. A. Lerman and B. Rabinovich 
Edited by M. Loeb 



Printed in Jerusalem by S. Monson 
Binding: Wiener Bindery Ltd. Jerusalem 



Available from the 
U.S. DEPARTMENT OF COMMERCE 
Clearinghouse for Federal Scientific and Technical Information 
Springfield. Va. 22151 



IX/ 9/ 5 



TABLE OF CONTENTS 



English page 



Preface 1 

Introduction 3 

Part One. PALEONTOLOGICAL AND ARCHAEOLOGICAL 

BACKGROUND OF FAUNA FORMATION 
Chapter I. Development of Caucasian Landscapes and Mammalian 

Fauna in the Tertiary 21 

Chapter II. Development of Caucasian Landscapes and Mammalian 

Fauna in the Quaternary 73 

Part Two. ANALYSIS OF THE ORIGIN OF CAUCASIAN 

QUATERNARY MAMMALS IN RELATION TO THEIR 

DISTRIBUTION, ECOLOGY, AND MORPHOGENESIS 
Chapter III. Review of the Early and Contemporary Populations and 

Ranges of Some Mammals 245 

Chapter IV. Stratigraphic and Geographic Variation in Caucasian 

Quaternary Mammals 384 

Part Three. GEOGRAPHIC ZONA TION AND LATEST EVOLUTIONARY 

TRENDS OF MAMMALIAN FAUNA OF THE CAUCASIAN ISTHMUS 
Chapter V. Analysis of Holocene Mammalian Fauna and Scheme 

of Zonation 453 

Chapter VI. Anthropogenic Changes in Mammalian Ecological 

Assemblages and Ranges in Various Zones 518 

Conclusion 587 

Bibliography 591 

List of Abbreviations 686 

Supplement. Part 1. Yield and Supply of Animal Pelts on the 

Caucasian Isthmus from 1925 to 1955 (Graphs 1-23) 691 

Supplement. Part 2. Maps of Animal Distribution on the Caucasus 

(Maps 1-97) 705 

Index of Common Names of Mammals 802 

Index of Latin Names of Mammals 811 



PREFACE 

N.K. Vereshchagin's book is somewhat unusual: although it is not 
a comprehensive treatise of the past and present evolution of the fauna of 
this interesting and complex area, he has succeeded in producing the first 
generalized and complete picture of the development of terrestrial fauna in 
the Caucasus from Neogene to Recent times against a background of the 
geological history and landscape shifts. 

In his work the author draws on the entire body of biological science, 
from paleontological data on individual stages in the phylogeny of Quaternary 
mammals to the voluminous data of modern zoogeographical and ecological 
conditions of contemporary forms. 

The author believes that the eastern Mediterranean (including the 
Caucasus) was characterized in the Neogene by a unique process in the 
mammalian evolution — faunal complexes replacing one another, as 
commonly occurred throughout the Recent in the Holarctic region. 

The Quaternary mammalian faunas of the Caucasus have their roots deep 
in the Pliocene. The Caucasian faunas proper of the mountain forest 
community evolved locally since at least Upper Miocene time. However, 
in the Quaternary the evolution of mammals in the Caucasus took place 
primarily through species invasions and extinctions. 

The evolution of new species on the Caucasian Isthmus was not as 
significant as other evolutionary processes in fauna formation. 

Vereshchagin's statements on the absence of saltations in the 
morphological evolution of the Quaternary mammals will require additional 
study before they are confirmed. 

Emphasis is placed on the destructive activity of humans which caused 
the extinction of many game mammals from the Upper Paleolithic to the 
Recent. Many examples are given of the ecological changes and 
readjustments of the surviving mammals in various zones. 

This work on the many questions relating to the development of fauna 
will undoubtedly inspire other intensive paleontological and zoological 
studies, not only on the Caucasus, but also on other parts of Russia. As an 
initiator, Vereshchagin has made a notable contribution to the methodology 
of his subject. 

The book also presents important data on the commercial value of 
autochthonous and introduced species, as well as information of concern 
to conservationists. 

The history of the development of mammalian fauna in the Caucasus will 
undoubtedly be of interest to a wide audience of botanists, geologists, 
archaeologists and geographers, as well as to zoologists. 

E.N. Pavlovskii, Academician, 
Lieutenant General, Medical Corps, 



INTRODUCTION 

A study of the distribution of organisms in time and space in the course 
of their evolution presents problems at once interesting and difficult. Their 
study is necessary for the reconstruction of faunal history and the 
advancement of faunalogical investigations. 

The study of the development of fauna on the Caucasian Isthmus is 
complicated by the fact that throughout the Cenozoic, this area was the 
scene of marine transgressions and regressions, mountain-building 
processes and glaciations. As a result of these processes, the climate and 
landscape changed within short distances. 

In the physiographic sense, the Caucasus is large and varied. Its geologic 
structures, its flora and fauna, and the history of its peoples have been 
studied by outstanding scientists for almost 250 years. 

Studies of the mammalian fauna of the Caucasus have been primarily 
concerned with the identification and geographic distribution of species. 
Few workers were concerned with the ecology, and even fewer with the 
history of the fauna. The main works are those of K. A, Satunin, 
N.Ya. Dinnik, N.A. Smirnov, M. V. Shidlovskii, Z.S. Rodionov, 
P. A. Sviridenko, S.I. Ognev, L. B. Beme, S.S. Turov, V. G. Geptner, 
A.I. Argiropulo, A. A. Nasimovich, S.K. Dal', I. V. Zharkov and 
P. P. Gambaryan. 

The vast collection of data by Russian zoologists on Caucasian animals 
can only be partly reflected in the bibliography of this book. 

Much paleontological work on the fossil mammals of the Caucasus 
has been done by V. V. Bogachev, A. A. Borisyak, V.I. Gromov, 
E.I. Belyaeva, V.I. Gromova, R.D. Dzhafarov, L. K. Gabuniya, 
N.O. Burchak-Abramovich, myself and others. The osteological collections 
of archaeologists (S.N. Zamyatnin, G.K. Nioradze, A. P. Kruglov, 
E.I. Krupnov and others) were of particular value. 

These studies and collections facilitated my investigations of the origin 
and history of the development of Caucasian mammals. 



The history of the mammalian fauna of the Caucasus is intimately related 
to the evolution of landscapes, vital forms and ranges of species distribution. 

The marine transgressions and regressions during the Tertiary in the 
eastern part of the Mediterranean geosyncline (the Recent Black Sea, 
Caspian Sea and the Caucasus) and the evolution of the molluscan fauna of 
this area have been studied in great detail by geologists: Academicians 
Andrusov (1888) and Arkhangel'skii (1934), Arkhangel'skii and Strakhov 
(1938), Zhizhchenko, Kolesnikov, Eberzin (1940) and others, mainly in 
the course of petroleum explorations. TheNeogene, in particular, has been 



thoroughly studied. A detailed zoogeographic summary on the faunas of the 
Caspian and Black seas was published by Sovinskii (1904). 

Geobotanists and paleobotanists have also published a number of histories 
of the land flora of the Caspian lands, particularly of the Caucasus 
(N.I. Kuznetsov, 1909; Palibin, 1936; Grossgeim, 1936, 1948; Maleev, 
1941; and others). 

Studies to date on the history of Caucasian land fauna have been 
quite generalized and were based mostly on fossil material. These studies 
include Academician Menzbir's (1934) history of the fauna of the European 
U. S. S. R., Serebrovskii's (1935) history of the fauna of the U. S. S. R. , 
Bogachev's (1938) review of the Tertiary of the Caucasus, and Borisyak 
and Belyaeva's (1948) review of the Tertiary mammalian fossil localities 
in the Caucasus. The history of the Quaternary fauna has been treated by 
Gromov (1939, 1948), Gromova (1948), Burchak-Abramovich (1951c) and 
Pidoplichko (1951, 1954). 

The history of the land mammals of the Caucasus has also been treated 
from the point of view of zoogeographical studies of the Recent by Satunin 
(1896, 1901a, b, 1904, 1909, 1913), Dinnik (191 1 ), and Shidlovskii (1940a, 
1941b, 1945, 1947), and more recently by B. A. Kuznetsov (1949, 1950). 

There are fewer studies on the evolution of other classes of terrestrial 
vertebrates: birds (Puzanov, 1938b) and reptiles and amphibians (Nikol'skii, 
1913; Sobolevskii, 1929; Lyaister, 1931; Chernov, 1939). 

The history of terrestrial invertebrates on the Caucasus has been 
discussed in the studies of scorpions by Byalynitskii-Birulya (1917)*; of 
dragonflies by Bartenev (1933, 1934 — in which he also discussed mammals, 
reptiles and amphibians); of Coleoptera by Semenov-Tyan-Shanskii (1936) 
and Bogachev (1947); and of Orthoptera by Uvarov (1921). 

All of these studies have one feature in common: the history of the fauna 
is treated either from a purely paleontological or from a purely 
zoogeographical point of view, without synthesis or consideration of the 
ecological and morphophysiological data. 

Obviously, this approach cannot give a true picture of a process as 
complex as the development of fauna at any given stage of geologic time. 
It is more correct to view the process of the development of fauna as 
a combination of three interrelated processes: 

1. Evolution of the environment and of those life conditions of individual 
species which were controlled by secular changes in climate, in elevation 
and subsidence of the earth's crust, and of the biocenosis proper. In more 
recent times the human factor also assumes great importance. 

2. Morphological evolution of individual species, which is affected by 
both external environment and internal developmental patterns. 

3. Ecological evolution of individual species and of entire biocenoses 
which is caused by continuous evolution of the conditions of life, and the 
further evolution of the biocenoses, in turn, through extinctions, 
transformations, invasions and migrations of species. 

Data on these three processes of faunal evolution are not equally 
available to paleontologists, zoogeographers and ecologists. A synthesis 
of the data and conclusions of each discipline is indispensable. 

The Caucasus has had a complicated geological, pedological and floral 
history, and a long pattern of changing human cultures. The influence of 

' [Arthrogastran Arachnids of Caucasia: Scorpions. Translated into English by IPS T in 1964, OTS No. 64-1114.] 



man on the land and the fauna became progressively more pronounced with 
time. Therefore, in order to understand the basic features of the evolution 
of Caucasian fauna, it is necessary to draw on the data of geology, 
geomorphology, geobotany, archaeology, paleontology, systematics, ecology 
and zoogeography. 

The complexity of the problem required a limitation on our studies and 
an organization into three interrelated subdivisions: 

1. Paleoniological and archaeological evidence of the history of the 
development of the fauna. 

2. Analysis of the origins and distribution of the Quaternary mammals 
with reference to their ecology and morphogenesis. 

3. The mammalian geography of the Caucasian Isthmus and the most 
recent manifestations of local faunal evolution. 

In each of the subdivisions, progress of the work depended upon the 
material and technical facilities available. 



Materials, routes and methods 

The main source of mammalian material of the Tertiary and Pleistocene 
is in the paleontological, geological and archaeological collections of: 
Museum of Natural History of the Academy of Sciences of the Azerbaidzhan 
S.S.R.; Zoological (ZIN) and Paleontological (PIN) Institutes of the Academy 
of Sciences of the U.S.S.R.; All-Union Geological Institute, Institute of 
Geology of the Academy of Sciences of the Georgian S. S. R. ; Institute of 
Geology of the Acadenny of Sciences of the Armenian S . S . R . ; and the city 
museums of Pyatigorsk, Stavropol, Krasnodar, Temryuk, Novocherkassk, 
Rostov and Astrakhan. Our own collections and observations on Tertiary 
mammals were taken from the Miocene site near Belomechetskaya in 
central Ciscaucasia, from the Pliocene site at the Kosyakin quarry near 
Stavropol and from the Upper Pliocene and Lower Quaternary sites along 
the Psekups and on the Taman Peninsula (Figure l). Collections of 
Pleistocene mammals and plants were taken from Pleistocene river sands 
in the Kuban and the Kuma river valleys, near Krapotkin and Georgievsk 
in Ciscaucasia, and from the bituminous formations of the Apsheron 
Peninsula in Transcaucasia (Figure 2). 

Small collections of bones and flint tools and additional observations were 
made at old excavation sites in paleolithic caves on the Black Sea coast 
and in the Rion River basin (Akhshtyrskaya, Sakazhia and Gvardzhilas caves) 
and in the recently excavated Kudaro cave. 

More than 55,000 Pleistocene bone fragments and 1,500 Tertiary bone 
fragments were examined. 

In order to trace the distribution of game animals and small rodents 
in the Holocene, the following types of fossil material from a variety of 
sites were studied: 

1. Food and industrial wastes of man from strata containing remains 
of campsites and settlements from the Mesolithic to the last centuries of 
the Recent (Sarkel, the ancient town of Semibratnoe, Elizavetovskoe, 
Cepi, Phanagoria, Taman, Gelendzhik, Anaklia, Kalakent, Mingechaur, 
Baku, and others). 



(9) 




2. Bone artifacts, ornamental objects, skulls and skeletons of wild and 
domestic animals from ritual burial grounds (Nal'chik, Samtavro, Trialeti, 
Sevan and others). 
9 3. Skulls of artiodactyls from mountain caves and shelters where they 

had accumulated over the centuries as the residue of rites and feasts 
dedicated to the hunting god (a number of ravines in Ocsetia). 

4. Bones of rodents and insectivores deposited by owls during the last 
centuries in caves, under ledges and in the small cavities of rocks (the 
foothills and mountain regions of Cis- and Transcaucasia). 

5. Bones and horns of drowned animals and kitchen middens of tribes 
which lived in pile dwellings along the shores of Lake Sevan. (The lake is 
at present receding rapidly. ) 

6. Isolated occurrences of bones in Holocene loams. 

In all, nearly 52,000 identifiable bones and bone fragments from 
approximately 70 Holocene sites (Figure 3) were studied. 

Penetration of species in postglacial times and their role in the evolution 
of faunal complexes were the primary considerations in our selection of 
smaller areas of Holocene localities such as the Pleistocene localities of 
the Apsheron Peninsula, Stavropol Plateau and Pyatigor'e area 
(Vereshchagin, 1949c, 1953a). We particularly searched out food rests of 
eagle owls . 

Mapping of the fossil localities and searches in the extensive literature 
10 produced a general picture of the distribution of bone -bearing deposits of 
the Tertiary and Quaternary. 

The elevation of the Caucasus in the Cenozoic resulted in an accretion 
of very thick strata of gravels, sands and silts in the piedmont plains. Part 
of the terrigenous material was reworked by waves in the surf zone and 
part was deposited in the quiet water of bays on the margins of sea basins. 

Accumulation and burial of skeletons occurred mainly along the deltas 
and coastal bays, where they were carried by streams. Cenozoic bone- 
bearing sandy-gravelly lenses are usually found in erosional channels cut 
through older beds; this type of burial is common in the valleys of 
Ciscaucasia, Stavropol and in the broad intermontane valleys of 
Transcaucasia. The bone -bearing formations are exposed in the processes 
of erosion and quarrying. 

Caves containing bones of Tertiary age have not yet been found in the 
Caucasus. This is probably accounted for by the changes in the relief and 
river network which destroyed the older karst. 

Archaeologists have searched for Paleolithic localities in the regions of 
11 developed Quaternary karst. The Paleolithic collections usually contain 
"fauna", i.e. , bones left in the caves by prehistoric hunters and by 
predatory animals and birds. The latest, post-Paleolithic (post-Pleistocene) 
bones occur mainly in surface loams, diluvial gulleys, alluvial deposits 
in the first terraces of rivers, in caves and under rocky ledges. 

The positioning of most Holocene bone -bearing burials accessible to 
excavation and investigation was determined by early hunters and predatory 
birds. However, evidence of mass extinction of wild animals and 
preservation of their remains even in Recent times can be found in the 
Caucasus (Vereshchagin, 1951b; Figure 4). 

It is clear that the paleontological record of the origin of Caucasian 
mountain fauna (which is the background of this work) is not complete. 



О) 




The reasons for this are taphonomical, i.e., the lack of accessible areas of 
sediment accumulations in the highlands and the absence of caves containing 
Pliocene fossils. (See Efrem.ov; "Taphonomy" (Tafonomiya) 1950.-) 

A critical evaluation of the available paleontological material requires a 
complete account of burials, their peculiarities and the ways by which bone 
remnants and matter were preserved in them. Therefore, a section dealing 
with the transport pathways and the burial of fossils has been added to 
each of the chapters devoted to the regional stratigraphy of the localities. 

A correct evaluation of the modes of preservation leads to important 
conclusions regarding the ancient landscapes and the conditions of life and 
of death of organisms. However, this factor was often ignored by 
paleontologists and, as a result, they were led to erroneous interpretations, 
particularly in cases where fossils of varying ages occur at the same 
locality. 

In most instances, the paleontologist deals with a collection of fossil 
remains which reflects only the conditions peculiar to a particular locality, 
rather than the ecological associations of the entire ancient region. 

The degree of "universality" of a burial is related to the number of 
preservable species of animals and plants which inhabit the area and to 
the proximal degree to which the distribution of species and individual 
specimens in the dead assemblage resembles that of the living community. 

Three sets of conditions determine the degree of universality of a fossil 
assemblage: l) the circumstances of the animal's death; 2) the manner inwhich 
the remains accumulated in sediment; and 3) their "behavior" and 
preservation within the sediment. 

Clearly, any possibility of reconstructing ancient biocenoses and 
landscapes from fossil material is determined by the sets of conditions 
given above. 

The Binagady burial on the Apsheron Peninsula (Chapter II) is an example 
of a highly universal fossiliferous site. There a Pleistocene assemblage of 
plants, birds, insects, jerboas and rhinoceroses was preserved in thick 
layers of asphalt; this mode of preservation permits a highly reliable 
restoration of the ancient landscape. 

The situation is quite different when fossils are preserved in sediments of 
diluvial, river, lacustrine or marine origin, or at Paleolithic sites or in caves. 

Accumulation of bone material in river sediment results from river 
erosion of bone-bearing beds (Vereshchagin, 1953c), as well as from the 
occurrence of animal deaths in the mainstream and on the floodplain. Once 
in the mainstream, the bones are subject to mechanical reworking like 
gravel and pebbles and they are sorted and distributed according to specific 
gravity and size. 

Thus, a complex of animals which died in a river valley can be seen 
accurately only if the bones are collected from different facies, i.e., gravels, 
sands, silts. 

Available collections by earlier investigators were not, as a rule, made 
in this way and consequently reflect only the last stage of sorting, i.e. , 
excavation. 

Distribution of bones in different facies also occurs in lakes and marine 
bays to which they are carried by rivers. Burial in deltaic sediments is 
more common, since floating bodies of animals can be carried long 

* [The term "taphonomy, " meaning "the study of the fomnation of burials of fossils, plants and animals, " 
was introduced by Efremov in this work.] 

10 



distances by currents to sink later into the silty and shell-covered bottoms 
of quiet bays. 

Accumulation of bones in the Paleolithic and post-Paleolithic strata of 
camp sites, settlements and burial grounds was primarily dependent upon 
the particular hunting traditions and customs of local tribes. (Deposition 
by flowing water is rarely encountered in caves. ) Remains of men's food 
was often found in caves mixed with remains of the food of predatory animals 
(bears, wolves, hyenas, panthers) and predatory birds (eagle owls and little 
owls). Clearly, gross errors are possible if, in drawing conclusions, human 
hunting customs and animal behavior patterns are ignored, as was done in those 
investigations which are limited to a count of domestic and wild animals in strata 
of varying ages and cultures. For example, it is known that contemporary 
Caucasian hunters often carry 30-40 kg of wild goat, gazelle or swine meat 
over distances of 20-25 km on mountain trails. Caucasian goats can climb 
1,500-2,000 m in a day. We have observed eagle owls in the cave region of 
Imeretia f]ying frequently from caves in the foothills to seek prey on 
the subalpine meadows and attaining heights of more than 1,000 m in their 
ascents. 

Thus, reconstructions of the paleolandscape and of changes in the 
position of the snow line based on occurrences in caves of the remains 
of highland animals (goat and Prometheomys) can be made only if all 
the processes involved in the accumulation of bones are considered. 

Similarly, conclusions on the accumulation of index gama fauna of one 
area cannot be based on a comparison of the material from an unlike area, 
e.g., material from the Greek towns of the Taman Peninsula (Phanagoria, 
Cepi) compared with the kitchen middens of a small hunting and fishing 
village of Roman time found at the ancient site of Semibratnoe in the mouth 
of the Kuban River. In this case the size of the settlement and its 
topographic location would be much more important considerations than the 
occurrence and disappearance of game animals in the surrounding area. 
14 Quantitative data on the composition and size of burials in the Caucasus 
are given here for the first time whenever adequate material was available. 
These data contribute to an understanding of the development of tribal 
culture and economy and of the nature of the landscape. To take one 
example: unverified ancient writings have led some geologists to believe 
that dry valleys existed in Colchis in the first millennium B.C. This theory 
is disproved by the absence of horse remains and the abundance of boar 
remains in the Colchis burials, which indicate that the marshlands of the 
Eneolithic and Bronze Ages were -^.^ery similar to those of approximately 
50 years ago. 

The species composition and distribution of wild animals were also 
inferred from the drawings and sculptures of the Bronze and Iron ages. 
Particularly interesting are the bronze, silver and gold objects with animal 
representations collected from the burials of the Armenian Highland, the 
Kura River valley and the Trans-Kuban sloping plain (Tsalka, Samtavro, 
Maikop, Kellermes, etc.). If one takes into account the probable origins of 
these objects, the aesthetic criteria of their consumers (Urartus, 
Scythians, Kobanians) and the ecology of the depicted, the realistic art of 
these ancient craftsmen can be of great aid in studying the distribution and 
gradual extinction of some of some of the larger animals. 



11 




"Ш^^р^ 



Sea bays and river deltas 



FIGURE 4. Predominant circumstances of animal death and types of fossil burials in the Cenozoic of the 
Caucasian Isthmus 



12 



Although it is possible to find fossils of large animals of the present 
period, their number has decreased to an extent that makes this a negligible 
probability. Consequently, the zoologist who is called upon by archaeology 
to furnish morphological and zoogeographic data for the identification of 
depicted animals will attach great importance to the so-called "animal 
style" in the art of the first millennium B.C. 

We have reviewed the published data of Tolstoi and Kondakov (1889), 
Uvarova (1900), Kuftin (1940) and others on precious stones in the "animal 
style" from the Caucasian burials of the Kobanian, Khodzhalian, Scytho- 
Sarmatian and other cultures, as well as the collections of the Hermitage 
(Leningrad), the Georgian Museum (Tiflis) and the Historical Museum 
(Moscow). Our conclusions, as they relate to species and ranges, are 
discussed in Chapter III. The "animal style" of these Caucasian antiquities 
deserves a special zoological-ethnographical study in itself. 

In the study of the Cenozoic faunas of the Caucasus we used the 
geochronological subdivision of the Neogene of the eastern Mediterranean 
(Black and Caspian seas, Caucasus) as given by Academicians Andrusov 
(1918) and Pavlov (1925), with later additions by Kovalevskii (1933, 1936), 
Zhizhchenko, Kolesnikov and Eberzin (1940). 

We placed the Pliocene-Pleistocene boundary (in agreement with Pavlov, 
1936, and Vardanyants, 1948) at the end of the Apsheron stage. The 
boundary probably corresponds to the last stage of existence of the Taman- 
Psekups fauna in Ciscaucasia, which was originally discovered by Gromov 
(1948, 1950) and is described in Chapter II of this book. 
15 The Quaternary of the Caucasus is subdivided into the Pleistocene and 

Holocene. Everywhere on the Caucasus the development of faunal complexes 
during the Pleistocene extended over a fairly long period, from the Upper 
Apsheron to the end of the last glaciation on the Caucasus and the onset of 
the xerothermic period which marks the beginning of the Holocene. The 
Pleistocene is subdivided into the Lower, Middle and Upper Pleistocene, 
which correspond to the stratigraphic stages of the Caspian region: the 
Baku, Khazar andKhvalynsk, containing mammalian fauna known from the 
entire Russian plain. The subdivision of the Quaternary in the foothills of 
the Caucasian Isthmus and the correlation of the river terraces with cave 
deposits and marine terraces were established following the studies of 
Mirchink (1937b), Gromov (1948), Vardanyants (1948) and Nikolaev (1953). 

The archaeological chronology of the Caucasus is similar to the European 
(West Mediterranean) schematic chronology developed by Obermaier (1913), 
Osborn (1924) and Penk (1939). For date determinations of the Caucasian Paleo- 
lithic as guide lines to grouping mammalian ecological complexes, we used the 
papers of Zamyatnin (1950b, 1957) and other archaeologists. Geological dating 
was a primary concern in our studies of bone material. 

The problem of determining the relative and absolute age of fossil bones 
is becoming increasingly significant for geology, paleontology and 
archaeology. Fossil age is determined by: l) physicochemical composition 
(the methods of calcination and radiocarbon dating); 2) taxonomic characters 
(generic, specific, subspecific); 3) inclusion in one faunal complex or 
another (faunal assemblages); 4) artifacts occurring with the fossils (stone 
tools, pottery, etc. ); 5) evidence of associated invertebrate fauna and 
indirect geomorphological data (for example, the age of marine and river 
terraces); 6) type of preservation of the bone material. 



13 



Each of the methods mentioned has disadvantages. The physicochemical 
methods give comparable results only when all the bones studied are less 
than 500,000 years old and have been preserved under the same conditions. 
The factors which affect the data obtained by this method are the composition 
of the sediments and their radioactivity, the water-salt relationship in the 
soil at the time of burial, etc. 

However, the method of calcination and comparison of the coefficients 
of the organic residue used by Pidoplichko (1952) can be used in absolute 
and relative dating provided that separate chronological scales are 
constructed for each physicogeographic region and for each type of locality. 

Radium-, uranium- and thorium-isotope dating methods (Cherdyntsev, 
1955) yield inconsistent results. 

Determination of geologic age by specific and faunal characters is 
16 possible only if the evolution of morphological characteristics is known in 
detail for the given genus or species. The taxonomic method can only be 
used with extremely well-preserved material, particularly in the case of 
skull specimens. 

Stone tools, pottery and other artifacts are reliable indicators of the 
relative age of fossiliferous strata only in those geographically defined 
areas where the chronology of the human cultures is known. By the latest 
consensus, the development of the Paleolithic cultures in Western Europe 
did not coincide in time with the Paleolithic in Eastern Europe. The 
occurrence of the bones in situ is another prerequisite for successful 
application of this method. 

Determination of the age of the geologic formations is of little use in 
determining the age of bones in Quaternary sediments which were often 
redeposited. The problem of dating is usually reversed since the Quaternary 
beds are dated by the vertebrates found in them. 

This method uses the simplest organoleptic analysis to determine the 
mode of preservation of bone material. But it requires considerable 
experience for an unbiased judgment and only yields indexes of relative age. 

It was necessary to review the entire body of Quaternary paleontological 
material taken from the Caucasus, since the indiscriminate application of 
morphological and morphometric criteria by earlier investigators had only 
resulted in confusion and redundancy in lists of species. 

With careful consideration of the mode and conditions of preservation, 
the age of the bone material can be placed within the major subdivisions 
of the Quaternary. The loss of organic matter (fossilization) and the 
secondary permineralization of the bones must also be taken into account 
by observing the new saturation of the bone material by salts, the 
color of fresh fracture, the degree and depth of colorization and the smell 
of fresh bone when scraped with a knife. A table is given for the 
determination of the relative age of Quaternary bones (Vereshchagin and 
Gromov, 1953a). 

The geographic distribution of Recent species was studied, with 
interruptions, from 1935 to 1941 in Azerbaidzhan and Armenia, and from 
1945 to 1952 in Azerbaidzhan, Georgia, North Ossetia, Kabarda, Svanetia 
and Dagestan. Our collecting routes are shown on the map (Figure 5). 

Studies were made of the collections of ZIN, the Zoological Museum 
of Moscow University, The Georgian Museum and Institute of Zoology 
of the Georgian Academy of Sciences (collections to 1944), the 
Institute of Zoology of the Armenian Academy of Sciences (collections 
to 1943), the Institute of Zoology of the Azerbaidzhan Academy 

14 





и 




н 




D 


U-, 


о 


о 


й 


Рн 


(Л 


< 


й 


^ 


о 




щ 




и 




р 




< 




15 



(collections to 1948) and the Stavropol and Baku anti-plague stations 
(collections to 1945). A total of 13,500 excellent specimens of skins, stuffed 
animals, skulls and specimens preserved in alcohol (taken from over 
3,300 locations) comprised the material used. 

Conclusions on the most recent changes in the density of 23 species of 
fur-bearing animals were based on the data collected over 25 years (1925- 
1950) by the governmental fur stations in Ciscaucasia and Transcaucasia. 
lY The data of 257 pelt-receiving centers for the period 1945-1948 were used 
in the final refinement of the areas of distribution of species, as shown in 
Figure 5. The statistics on the preparation of furs are shown separately on 
the graphs for Ciscaucasia and Transcaucasia, and for the republics and 
regions (except the Astrakhan Region, for which no data were available). 

The occurrences of fossils and the recent yield of small fur-bearing 
animals are entered on a specially prepared map drawn to a scale of 
1:5,000,000. The population density of each species, as estimated from the 
mean annual number of furs received by the pelt-receiving centers, are 
indicated by small dots on the map. The total number of pelts exceeds 
3,665,000. Range boundaries have been drawn only for stenotopic species 
(i.e. , those with a narrow geographic range). The occurrences of widely 
distributed animals, such as boar, gazelle, deer, antelope, saiga, goat and 
otter, are given according to our own observations. 

Data from the literature, not confirmed by observation, are entered 
with a query. 

The history of the development and the present state of distribution are 
given only for the stenotopic species and for those which are well 
represented in the paleontological record. 

The history of animal distribution in the Caucasus is closely connected 
with the climatic zones of the Quaternary. A relationship between vegetation 
and species distribution is also valid for this period during which there 
was a successive displacement of xerophytic desert vegetation, first by 
a steppe landscape, then by a mesophytic, and lastly by a forest landscape 
followed by a reversal of the entire process of vegetation displacement. 
The relationship, however, is not as clear in the older periods. It is certain 
that conclusions on distribution areas are easier to reach and are more 
reliable, as the species association with a given type of vegetation and 
landscape is longer and its ecologic niche is more limited. 

Gaps in distribution and the influence of ecological barriers are only 
considered as they relate to types of vegetation and to barriers which were 
geophysical and climatic in nature. Ecological barriers which are 
determined by physiological reaction norms and morphological adaptations 
of the species require additional study. The validity of our theoretical 
assumptions was confirmed by the finding of the indicator species 
associated with a given type of vegetation and landscape in every case. 

The morphology was studied mainly from the Quaternary fossil material in 
relationship to its stratigraphic distribution and the rate of evolution of the spe- 
cies. The morphological studies were primarily of carnivores (Vereshchagin, 
1951b) and of some rodents and hoofed species (Chapter IV). The geographic 
variability of some Recent carnivores, ungulates and rodents was also 
1° investigated in order to understand their origins and geologic age. 



16 



19 



The taxonomic classification of fragmentary bones of Quaternary 
mammals proved to be a problem. Some mammals, known from the Middle 
Pleistocene to the Recent, show gradually evolving successive forms with 
very few morphological differences. Other lineages of mammals are 
represented by a succession of more or less distinguishable forms which 
can be discussed as subspecies of the Recent or fossil species, or as species. 

As a rule, paleontologists either classify the Upper Quaternary forms 
as subspecies of existing species, or they identify the fossil with the Recent 
species. This identification of a fossil form as a subspecies of a Recent 
species is unsound, as the criteria for subspecies distinction are quite often 
only conventional and subjective. 

Although we are out of agreement with this practice of classification, 
we retained in Tables 62 and 103 the names of the subspecies found in the 
literature. Those subspecies whose stratigraphic records are inadequate 
are entered with the qualifications conformis (similar) and affinis 
(related). Unfortunately, the Rules of Zoological Nomenclature (1932) 
do not cover this problem. 

Wherever possible, we applied ecological data of the Recent to the past, 
and the study of the "universal" death assemblage at Binagady proved the 
effectiveness of this approach. 

Although a certain constancy of predominant features is recognizable 
in Quaternary organisms, it is probable that there was a fairly rapid and 
extensive evolution of the physiology. This brought about modifications in 
the mode of life without changing the former morphological features. 

The zoogeographical arialysis of the entire fauna of the Holocene mammals 
of the Caucasus is the culmination of studies of the history of faunal 
complexes. The analysis is based on genetic and stratigraphic principles. 
Rather than adhering to geomorphological and phytogeographical units, we 
have emphasized the evolutionary aspects of areas of species distribution 
and of faunal complexes. 

The problem of the geographic origins of species is discussed in 
Chapter III along the lines of criteria proposed by Arldt (1919). 

In any discussion of the future evolution of ranges, the chief consideration 
for any geological stage is undoubtedly the human factor. With this in 
mind, we studied the numbers and behavior of animals in their original 
biotopes and in those influenced by man, selecting various geographic zones 
mainly in the eastern Caucasus. 

These investigations add to our understanding of recent evolutionary 
development or degradation and direction of the fauna. 

Quantitative studies of animal populations in nature preserves were 
carried out during the expeditions (1938-1948) of the Azerbaidzhan anti- 
plague station and the Institute of Zoology of the Academy of Sciences of 
the Azerbaidzhan S.S.R. Most of the studies employed commonly known 
methods, although in some cases special new methods nad to be devised. 

Studies of recent changes in virgin steppelands were done mainly in 
Azerbaidzhan and should be carried out in the future in Ciscaucasia where 
shelterbelts have been planted. 

Introduction by man of new species into the existing fauna is not 
discussed in detail in this book. Acclimatization of new species and the 
biotechnical procedures of game farming, which have been studied by many 
scientists, are separate problems. Nevertheless, some examples are given 



17 



in this book of the rates and patterns of distribution of some introduced 
carnivores and rodents. * 

The effects of man's alterations of the landscape were clearly observable 
in the most recent changes in ecological assemblages; in fact, they may 
completely overtake the natural processes of change. For example, the 
development of stunted thorny shrubs in the place of forest in the foothills 
of eastern Ciscaucasia is a result not only of the Recent climatic trend, but 
also of deforestation and cattle grazing. Another example is the artificial 
drainage and forestation of the central parts of the Colchis Plain, which 
completely suppresses the development of water-logged soils and hydrophilic 
vegetation. In both cases the original assemblages of large and small mammals. 
were completely destroyed. 

The following ecological and zoogeographical terms are used throughout 
the book: 

Ecological assemblage of mammals — a group of species which inhabit 
one biotope. The morphological and physiological features of the species 
are the result of evolution within the framework of the existing ecological 
conditions . 

Faunal complex — a number of ecological assemblages occurring within 
one homogeneous geographic zone. 
20 For example, the faunal complex of the arid eastern Transcaucasian 

plains from the Pleistocene to the Recent includes ecological assemblages 
of semidesert, tugai** vegetation, reed- grown lakes and swamps and other 
types. This complex has evolved since the Pliocene under dry and 
moderately warm climatic conditions. 

The term 'fauna' is used in the book to designate a number of faunal 
complexes occurring in a multizonal territory. The word 'fauna' is 
commonly used by geologists, paleontologists and archaeologists to designate 
any collection of fossils. We do not use the word in this sense: where it was 
necessary to employ it, it appears in quotation marks ("fauna"). It is a 
mistake to identify a selected collection of fossil remains of animals, 
possibly of different ages, with the fauna, or the faunal complex, of a given 
region or country. 

Our terms are quite applicable to paleogeography, i.e. , to the older 
stratigraphic stages. A faunal complex can exist throughout an entire 
geological epoch, during which its composition will gradually evolve. The 
extent to which it is possible to reconstruct fossil ecological assemblages 
and faunal complexes depends upon the degree of completeness of the 
geological record. It is for this reason that index species which occur in large 
populations at certain stratigraphic horizons are indispensable to a 
reconstruction . 

From the occurrences of such index species of the Pleistocene steppes 
of Eurasia as mammoth, horse, bison, saiga and other species, it is 
possible to infer the occurrence also of suslik, marmot, corsac fox, 
Siberian polecat, steppe skunk and related species. Similar inferences 
can be made from the occurrences of mountainous index species. 

In the process of preparation of this book, it was necessary to use 
extensive paleontological, zoological, botanical, geological and 
archaeological literature. Of the more than 3,200 sources consulted, only 
the most frequently quoted are given in the bibliography. 

' The author participated in the introduction of nutria, common raccoon and mink into the Caucasian fauna 

and served as a consultant in a number of other experiments. 
'* [Tugai —a bottomland complex with forests, bushes and meadows in river valleys of Central Asia.] 

18 



The writings of ancient Greek, Ronnan, Arab and Armenian naturalists, 
historians and geographers in translation, as v;ell as the reviews of Gan 
(1884-1890), Latyshev (1893-1904, 1947, 1948), and Karaulov (1901 ) 
were used as source material. 

Of the voluminous literature on species systematics only the titles most 
essential to the Caucasian fauna are given. 

The Caucasus (Caucasian Isthmus) is considered to lie within the 
following boundaries: the Kuma and Manych rivers to the north, and the 
international boundary of the Trancaucasian republics to the south. 

Of the many questions and problems which arose during the course of 
the field work and preparation of the book, some of the most important 
21 involved the stratigraphic correlations of paleolithic localities in mountainous 
country wherein several climatic, floral and faunal provinces are joined. 

The present work could have been accomplished only with the support 
of the Directors of the Zoological Institute of the Academy of Sciences of 
the U.S.S.R., and the Presidium of the Azerbaidzhan Academy of Sciences . 
The author is indebted to B.S. Vinogradov and A. A. Strelkov for valuable 
editorial comments and advice. 

The line drawings were done by V.N. Lyakhov, the drawings of animals 
by E. Ya. Zakharov and Prof. K.K. Flerov. All the photographs were taken 
by the author. 

The interest and cooperation of official institutions and individuals, both 
specialists and laymen, was invaluable to the field work in Azerbaidzhan, 
Armenia, Georgia, Dagestan, North Ossetia, Stavropol and Krasnodar. 

I thank all those who, in one way or another, contributed to the 
completion of my work. 

The first and more detailed manuscript was completed in 1954; in the 
process of preparation for this publication, many sections were shortened 
and new material was included. 



19 



25 Part One 

PALEONTOLOGICAL AND ARCHAEOLOGICAL 
BACKGROUND OF FAUNA FORMATION 



Chapter I 

DEVELOPMENT OF CAUCASIAN LANDSCAPES AND 
MAMMALIAN FAUNA IN THE TERTIARY 

TERTIARY BONE-BEARING LOCALITIES 

Abundant remains of Carboniferous plants in the slates and sandstones 
along the northern slopes of the Main Range indicate that land existed in 
the Caucasus as far back as the Paleozoic. 

Individual islands appeared after the broad transgressions of the 
Mesozoic, and later merged gradually into one landmass. 




FIGURE 6. Land formations (cross-hatched) in Eastern Europe 
and Southwest Asia (from Arkhangel'skii and Strakhov, 1938) 
Dot indicates Upper Oligocene locality at Benara 



The Caucasian islands were covered by ferns, ginkgos and cycads in 
the Jurassic, but the paleobotanical record only commences with the 
Cretaceous when plane, poplar, dryandra, myrica and giant conifers 
(araucarias, sequoias and others) appeared. 

Iguanodons which occur in travertine encrustations in the Satapliya 
cave near Kutaisi are the only fossils representative of the terrestrial 
vertebrates of the islands (Kandelaki and Dzontsenidze, 1937; Gabuniya, 
1956b). 

According to Pavlov (1936), the area between southwest Asia and the 
Baltic was dry land in the Paleocene. The Caucasus was built of gently 
folded Jurassic and Cretaceous strata. 



21 



At a later stage, the transgression of Tethys covered southern 
Europe, the Ukraine, the Crimea and part of the Caucasus, and Tethys 
became connected with the Asian sea extending to the southeast. 

Thus, the formation of land in the Caucasus, which determiined the 
evolution of the Quaternary faunas, took place entirely during the Cenozoic. 
26 In the Eocene the northern coast of Asia Minor, Surami, Kazbek and 

Elburz were sites of intensive vulcanism, which probably produced the 
high relief of the island lands of the Caucasus (Figure 6). There are no 
data on the Eocene flora. 

Small whales of the genus Zeuglodon (Archaeoceti) lived in the sea 
surrounding the island of Caucasus. Their fossilized remains occur in the 
Koun beds of the Apsheron Peninsula and in the area of the Sunngait rivulet. 

The first occurrences of terrestrial vertebrates in the Caucasus are 
in the Oligocene. 



OLIGOCENE 

The tropical character of the Caucasian landscape persisted through 
the Oligocene. The tuffaceous sandstones of Lower Oligocene age in 
Mount Darry-Dag and the Araks valley contain ferns (Blechnum brauni, 
Pteris oe n i n ge n s i s), palms (S a b a 1 h e a r i n g i an a), conifers 
(Po d о с a r p us), Lauraceae (C i n n am о m u m), myrica, zelkova, and 
small grasses (Pan i cum miocenica and other forms). Palibin (1936) 
correlated the Darry-Dag Oligocene flora with the Lower Oligocene floras 
of the Balkans and central Europe. He thought that the Darry-Dag is 
indicative of humid tropical forests covering volcanic islands. 

No bones have as yet been found in the Lower Oligocene beds of the 
Caucasus. However, it is possible that the island of Caucasus and the 
adjacent land masses to the south were inhabited by the same fauna of 
primitive carnivores and hoofed mammals which are well-known from the 
Eocene and Oligocene deposits of Western Europe. 

Bogachev (I938d) mentions a footprint of a creodont about the size of a 
large dog from a layer of silicified volcanic ash in the Darry-Dag near 
the town of Dzhulfa. 

The marine (Middle Oligocene) Maikop beds of Cis- and Transcaucasia 
contain abundant remains of cod and herring. Occurrences of 
Halitherium cf. schinzi Каир, and other, unidentified, sirenian 
species (possibly the ancestors of dugongs) are known from the Chiaturi 
manganese basin of Georgia, on the southern coast of the Black Sea, and 
from the Maikop beds in the Akhzy-Khazry area of the Apsheron Peninsula 
(Bogachev, 1938c). The whales Zeuglodon (M i с r о z e u g lo d о n) 
caucasicum Lyd. and Iniops caucasicum Lyd. have been described 
from beds that are probably the same age (Lydekker, 1892). 

Remains of Microzeuglodon aff. caucasicum Lyd. have also 
been described from the beds eroded by the Sumgait rivulet 20 km north of 
Baku (Ryabinin, 1938). 

Occurrences of the whc^le genera Microzeuglodon, Iniops, 
Delphinus (sensu lato) and Zeuglodon have been recorded (Bogachev, 
1938c, 1939a) from the Maikop beds of the Apsheron Peninsula. 



22 



The same Maikop beds along the Sumgait rivulet contain fossil leaves of 
the evergreens Combretaceae, Sapotaceae, Ternstemiaceae, with an 
admixture of tropical conifers. 



(27) 








FIGURES 7. Bones of hoofed mammals from the Oligocene of Benara 

1-4 — РтзЯпс! OS lunatum of Benara theriu m callistrati Gab.; 2 — M^ of A с er a th e г iu m 
cf. filholi Osb. ; 3 — metatarsus of S ch izotheriu m chucua; 5 — fragment of scapula of small 
artiodactyl 



The abundance of marine mammals and the dense land vegetation attest 
to the earlier existence of tropical conditions on the island of Caucasus and 
of Sargasso-type lagoons on its southern coast. Observations on the 
27 southern slopes of the Taurus Mountains indicate that Asia Minor was also 
covered with such forms as podocarpus, myrica, oak, cinnamon, 
andromeda and eucalyptus in the Oligocene. 

A burial containing Oligocene land mammals was discovered by the 
geologist M. F. Khuchua in 1948 at Benara in the Akhaltsikhe region of 
southern Georgia. A few poorly preserved bone fragments and teeth were 
found in upper multi-colored sandstone, conformably overlying the lignite 
beds (Figure 7). The mode of preservation in situ was inadequately studied. 



1704 



23 



Small bone fragments occur mainly in the concretions made of hard 
sandstone. The material was identified and described by Gabuniya (1951a, b, 
1953, 1955a, b) as rodent and ungulate: 

Rodentia 

Fam. gen. 

Perissodactyla 

S с h i z о t h e r i u m chucua Gab. 

Be na га I he ri um callistrati Gab. 

Artiodactyla 

Anthracotherium sp. 
Lophiomeryx benarensis Gab. 

28 Gabuniya has correlated the Benara locality with the Oligocene 

phosphorite beds of Quercy in France and with the Middle Oligocene 
localities of Mongolia and Kazakhstan, where bones of Schizotherium 
and Lophiomeryx are known from the Upper Oligocene. 

The fauna from the vicinity of Akhaltsykh confirms the paleobotanical 
evidence of the land connections which existed, with interruptions, between 
the Caucasus and central Asia and between the Caucasus and Western 
Europe in the Oligocene. 



MIOCENE 

In the Lower Miocene the island of Caucasus extended from the vicinity 
of Anapa in the west to the upper reaches of the Samur River in the east. 
The area of the Dzirul crystalline massif was also an island of smaller 
dimensions. The region of Trialeti-Akhaltsikhe and Borzhomi was covered 
by sea. Sands were deposited in the sea, although toward the close of the 
Lower Miocene the area became dry land (Zhizhchenko, Kolesnikov and 
Eberzin, 1940). Only marine mammals are known from the Lower Miocene 
beds. 

Bones of the seal Phoca sp. , resembling P. vindobonensis Toula 
(Bogachev, 1927b), occur in the Miocene beds at Lok-Batan, south of Baku. 
Remains of Phoca sp. have also been reported from the limestones of 
the Kilyazinskaya spit north of Baku, and of the whale Cetotherium 
meyeri Brandt from Lok-Batan and the village of Dzhorat on the Apsheron 
Peninsula (Bogachev, 1938c). 
29 The geography of the island of Caucasus at the beginning of the Middle 

Miocene, in the so-called Helvetian stage, was essentially the same as in 
the Lower Miocene. The sea, in which Tarkhan and Chokrak sediments 
accumulated, covered the entire area from the northern coast of the island 
of Caucasus to Rostov in the northwest, and Stepnoi (formerly Elista) in 
the northeast. At that time the landmass of Caucasia Minor probably 
appeared south of Surami island, and the Caucasian island increased in size. 

The vegetation of the islands in the Middle Miocene, as shown by the incidence 
of fossils in Cis- and Transcaucasia, retained its subtropical character 
(Palibin, 1936) though with a small admixture of deciduous trees (Castanea 
s a t i va) and conifers (P i n u s n e p t un i) of northern origin. In the Middle 



24 




25 



Miocene, in Chokrak tinne, the Caucasian land gradually became higher and. 
in the process of growth, became a peninsula of the southwest Asia massif 
(Figure 8). 

The sediments of Chokrak age in Ciscaucasia contain abundant remains 
of terrestrial mammals. 



Central Ciscaucasia 

Be lome che t skay a. This locality, discovered by A. V. Danov in 
1926, is near the Cossack village of Belomechetskaya, on the right bank 
of the Kuban River, north of Cherkessk (Figure 1). 

In this region of the plateau, the Chokrak beds, 60 m thick, are mostly 
coarse-grained sandstones plus a combination of sand, clay and calcareous 
sediments. In a number of places, sandy bone-bearing lenses have been 
exposed in the Miocene hills by the eroding action of the Recent Kuban 
River approximately 50 m above the present water level. These were 
probably formed in a river which cut through the Chokrak beds eroding an 
ancient accumulation of skeletons in a lake or a marsh. The bone-bearing 
lense at the southern outskirts of Belomechetskaya is made up of gravelly 
quartz sand, with grains measuring 0.5-1.0 mm in diameter, slightly 
cemented by silt. Parts of the outcrop are iron-stained in vertical and 
diagonal bands. There is no pronounced bedding, which, however, may be 
a result of metamorphism. 

Areas of silty sandstone, more strongly cemented and of grayish color, 
occur within the mass of the gravelly quartz sand. The silty sand indicates 
changes in the river current regime. The exposed thickness of the bone- 
bearing layer at this locality is 3.5-4.0 m. It is overlain by a limestone 
layer 30-40 cm thick. This layer probably corresponds to a brief marine 
transgression. Higher in the section the sands alternate with limestones. 
In places the sands are covered by Quaternary loams (Figure 9). 

The remains of land and marine animals (bones and bone fragments, 
fragments of skulls, isolated teeth) are scattered throughout the layer. 
They occur in the gravelly sand and more often in the bluish, cemented, 
silty and iron-stained sand. The surface of the bones is pale yellow, light 
brown or dark brown in color. In fresh fracture the diaphyses are light 
brown, the cracks and the pores of the epiphyses are stained with ferric 
or, more rarely, ferrous oxide. The tooth dentine is also stained; the 
30 enamel is glossy and dark blue, or sometimes black at the base of the 

crown. The old factures on the long bones were rounded by water and the 
cavities in the bones were filled with sand. 

No tooth marks of predators were found on the bones. It is difficult 
to determine why and in what season of the year the animals died. It seems 
doubtful that death was caused by catastrophe. The variety and the 
fragmentation of the bones, and the abundance of teeth of small herbivores 
which occur sometimes in horizontal, sometimes in vertical positions 
within the sediment, indicate several cycles of redeposition. The death 
assemblage contains both young and old herbivores and carnivores. 
Remains of the swamp mastodon (Platybelodon) consist of a fragment 
of the facial part of the skull and mandibles which belonged to one adult 
and one young individual. 



26 





FIGURE 9. Bone-bearing bed at Belomechetskaya 

Photograph by author, 1950 



The bone material at this locality includes both terrestrial and marine 
mammals, and the aquatic turtle Trionyx. The mode of preservation of 
31 all the bones is very much the same. This indicates the redeposition of the 
bones in alluvium. The bones were probably washed from the skeleton- 
bearing lenses of Oligocene -Miocene age by rivers or, less likely, by sea 
waves. In this process the bones became sorted and redistributed in the 
sediments of the river bed. The first burial of the bodies of land animals, 
as indicated by the iron stain on the bones, took place in a basin of standing 
fresh water, probably on the floodplain. Thus, it is evident that the 
existence of the Belomechetskaya "fauna" antedates the deposition of most 
of the animals in the fossiliferous layer, although some few of the "fauna" 
undoubtedly died during the time of deposition. In this process the bodies 
of land animals carried by the river and the floating bodies of marine 
mammals entering the river mouth were macerated and their remains 
incorporated into the sediment. 



27 



The collections of PIN, ZIN and the Institute of Paleontology of the 
Georgian Academy of Sciences include the species and groups of mammals 
given in Table 1 . 



TABLE 1. Species and number of mammal bones from the Middle \4iocene beds at Belomechetskaya* 





Number 




Number 




of bones 




of bones 


Camivora 




Anchitherium aurelianense Cuv. 


4 


Amphicyon sp 


9 
6 
4 


Dicerorhinus caucasicus Boris. 
Chalicotheriidae gen 


172 




3 


Hyaena sp 


Fam. gen 


1 


Ursavus sp 


Artiodactyla 




Tubulidentata 




Orycteropus sp 


1 


К u ba no с hoe ru s robustus Gab. 
Hippopotamidae (?)gen 


1 
1 


Rodentia 




Micromeryx flourensianus Lartet 
Dicrocerus elegans Lartet 


7 
3 


Palaeocricetus caucasicus Arg. 


6 


Pa га tra gocerus caucasicus Soc, 


8 






Eotragus cf. martinianus Lartet 


12 


Proboscidea 




Hypsodontus miocenicus Soc. 


1 




4 


Fam. gen 


52 


Platybelodon danovi Boris 






Mastodon sp 


131 


Cetacea 




Perissodactyla 
Paranchitherium karpinskii Boris 


20 


Fam. gen 


34 







* Material identified with the aid of BorisyakC 1928a, 1943), Argiropulo (1938, 1940c), Sokolov (1949), 
Gabuniya (1955d, 1956a) and unpublished data of K.K. Flerov and of the author. 



The collections also include fish and aquatic turtle (Trionyx). The 
total number of known Miocene mammals from the Belomechetskaya is 
approximately 20, although the taxonomic identity of some specimens 
remains unknown (Figures 10-15). 

Paranchitherium, Anchitherium, Rhinoceros and 
Platybelodon were considered stratigraphic index fossils by Borisyak 
(1937, 1938b). The degree of specialization of Paranchitherium from 
32 Belomechetskaya is the same as that of Parahippus from the Miocene 
of North America. Anchitherium, which is probably an earlier form, 
is characteristic of the Miocene of Eurasia. 

The Belomechetskaya rhinoceros was considered by Borisyak as one 
of the most primitive Dicerorhininae. The dentition shows both primitive 
and specialized features. Possibly it should be regarded as the oldest 
representative of Rhinoceros schleiermacheri. 

A peculiar mastodon, Platybelodon, is closest to the genera 
Phi о mi a and Palaeomastodon of the Oligocene of Africa, which 
may possibly be the ancestors of the long-snouted mastodon. The lower 
tusks of Platybelodon, flattened dorsoventrally and firmly joined, 
form an elongated scoop similar in shape to the tusks of Amebelodon 
fricki Barb, from the Pliocene of Nebraska. Accordingto Borisyak (1928b) 



1704 



28 




FIGURE 10. Bones of carnivores from Belomechetskaya 
1, 2 — Hyaena sp. ; 3— Amphicyon sp. 



and Osborn (1936) these mastodons had no trunk. They lived on the shores 
of lakes and river backwaters, digging with their flattened tusks for the 
succulent water plants and roots which were their food. 

Small Dicrocerus, Micromeryx and Eotragus are the main 
index fossils of the Middle and Upper Miocene of Eurasia. In general the 
fossiliferous layer at Belomechetskaya contains mammals of stratigraphically 
different ages. In view of this, it is difficult to identify the index species 
of the Middle Miocene faunal complex. 
33 The peculiar features of the Belomechetskaya "fauna" do not permit 

an easy comparison with other Miocene faunas of Eurasia and therefore 
its zoogeographical relationships are not clear. However, since the 
autochthonous development of such a fauna on the small landmass of the 
Caucasus is unlikely, an earlier land connection between the Caucasus 
and the continental landmass in the south can be regarded as certain. 

A reconstruction of the Caucasian landforms of the time when the animals 
preserved at Belomechetskaya lived would include mountain ridges and 
34 broad valleys with forests, bushland, lakes and quiet rivers as part of the 
landscape. The backwaters of those rivers and lakes were .probably the 
feeding grounds of Platybelodon. 

Fossil insects and plants occur in the Middle Miocene beds which contain 
marine mollusks. These beds belong to the Karagan and Konka stages. 



29 



(33) 




FIGURE И. Skull and tooth of Platybelodon danovi (from Borisyak, 1928b) 



Poplars (Populus mutabilis), Sapindus fulcifolius and 
evergreen cinnamons (Cinnamomuna scheuchzeri) occur in the 
Karagan beds of Dagestan. The sandy-clayey littoral sediments of the 
Spanidontella sea, southwest of Derbent, contain leaves of evergreen species 
characteristic of the Upper Oligocene and Lower Miocene of Europe: 
Myrica hakeaefolia, Laurus primigenia, Ardisia cf. 
oceanica, Diospyros paradisiaca and other forms . The fauna of 
marine mollusks indicates that the Miocene sea became progressively less 



30 




г^ 



FIGURE 12. Skull and tooth of Para n chitherium karpinskii (from Borisyak, 1937) 



saline due to its isolation from the Tortonian basin of the Balkan region, 
that the climate became more humid and that the rivers carried more water 
(Zhizhchenko, Kolesnikov and Eberzin, 1940). 

The richest "fauna" of insects (some 90 identified species) was 
discovered in bedded marls of the Karagan stage west of Stavropol. The 
insects were probably carried by streams into the shallow lagoons of the 
35 northern coast and buried in the silty sediment. The following aquatic 

insects are common: mayflies (E phe m e r о p t e r a), dragonflies (Odonata), 
true bugs (Gerridae), water beetles (Dytiscidae, Hydrophilidae), caddis 
flies (Trichoptera) and mosquitos (Diptera). Among the terrestrial 
phytophagous, saprophagous, predaceous and parasitic forms are 
Orthoptera, Hemiptera, aphids, cockroaches, termites, cicadas, butterflies 
(Lepidoptera), Diptera, Hymenoptera and beetles (Rodendorf, 1939). 

This assemblage suggests a subtropical climate and profuse grassy 
vegetation in the Caucasus at the end of the Middle Miocene. 

The numerous plant fossils from the Krynka River valley in the Donets 
Basin (Krishtofovich, 1930) indicate that the vegetation on the northern 
shores of the Konka basin was of the temperate type and strikingly different 
from the subtropical flora of the Caucasus in the Neogene. 

In the Upper Miocene the seas still covered the southern Ukraine, the 
Crimea, Ciscaucasia, most of the Transcaspian land and possibly Iran. 



31 





FIGURE 13. Upper incisors of giant 
perissodactyl from Belomechetskaya 



FIGURE 14. Canine of an 
unknown artiodactyl 
(Hippopotamidae? ) from 
Belomechetskaya 



The fauna of the Sarmatian sea reflects a further decrease in salinity 
and shows considerable variability in facies and in stratigraphic age. 

In the Lower Sarmatian the Caucasian landmass grew southward. The 
thickness of the Sarmatian sediments indicates that the relief of Japhethida 
and of the coastal areas was essentially the same as in the Middle Miocene. 
The areas of maximum uplift in the Lower Sarmatian were in the Guri ridge 
and in Khevsuretia, and in the Middle Sarmatian, in inner Dagestan. The 

36 northwestern coast remained a plain. The sea in this area became shallower 
due to the uplift of the Stavropol massif. 

In the Upper Sarmatian the sea receded from the northern part of the 
Caspian and from most of the Transcaspian lands, leaving the Stavropol 
Plateau as a peninsula projecting into the Ciscaucasian strait. To the south, 
a wide isthmus extended from Kutaisi to Telavi connecting Japhethida with 
the Caucasia Minor landmass, and in eastern Transcaucasia a narrow Kura 
bay formed (Zhizhchenko, Kolesnikov and Eberzin, 1940). 

The landscapes of the Caucasus in Sarmatian time can be restored with 
a high degree of certainty from the available plant and animal fossil material. 

37 The littoral marine flora consisted of species of Fucus and 
Cystoseira, which occur as fossils in the valleys of the Sunzha and the 
Sulak on the Malokabardinskii ridge. 

Seals (Phocidae) and small toothless whales of the genus Cetotherium 
were abundant in the Sarmatian sea. Bones of Cetotherium have been 
found in the Sarmatian coquinas and silts near Derbent, in Makhachkala, 



32 



and in the blue Sarmatian clays near Maikop (Spasskii, 1939). Bones of seal 
have been found in the white marine sands near Stavropol, in the blue clays 
near Maikop and Goryachi Klyuch and in a number of other places. The 
Stavropol seal bones were found with bones of terrestrial Tertiary mammals 
— undoubtedly species which lived in a warm climate. 



(36) 




FIGURE 15. 

1-jawof Kubanochoerus robustus (from Gabuniya, 1955a); horn axes of: 2 - Para tragocerus 
caucasicus; 3-Eotragus cf. martinianus; 4 - teeth of Hypsodontus miocenicus (from 
Sokolov, 1949) 



The terrestrial flora of the Caucasus in the Sarmatian is represented 
by coniferous, evergreen and deciduous species. Remains of spruce 



33 



(Abies sp. )*, pines (Pinus sp. ) and sequoia (Sequoia sp. ) occur in 
the exposures along the Supsa River. Magnolia, laurel (Laurus 
p r i m i ge n i u s), cinnamon (Cinnamomum po 1 у m о r ph um), elm, 
willow and pear trees have been identified from the Kakhetian Range and the 
Trans-Kuban Plain. 




FIGURE 16. Land formations in the Upper Miocene (Upper Sarmatian) 
(From Kolesnikov; see Zhizhchenko and others, 1940,) 
Dots indicate main occurrences of land vertebrates 

At some of the Middle Sarmatian localities in eastern Georgia as many 
as 30 arboreal species have been collected. Of these, up to 70% are 
deciduous and include the Tertiary species of hornbeam (Carpinus), oak 
(Que reus), willow (Salix), maple (Acer), chestnut (Castanea), 
zelkova (Zelkova ungeri), and shrubs of Indian hemp (Apocynum), 
bladder senna (Colutea salteri) and other deciduous species in addition 
to the evergreen laurel, cinnamon, nayrica and magnolia. According to 
38 Palibin (1936) the evergreen species of the flora "which covered the entire 
surface of Japhethida" were the last Sarmatian representatives of the 
subtropical and tropical flora which flourished on the Caucasus from the 
beginning of the Tertiary. This first step in the borealization of the flora 
contributes to an understanding of the later evolution of the fauna. In a 
comparison of the Caucasian flora with the Upper Miocene flora of Asia 
Minor, Grossgeim (1936) found that they are very similar, both being 
characterized by a mixture of boreal and subtropical elements. 

The taphonomy of these dead plant assemblages is significant for the 
ecologist or faunist in the study of altitudinal zonation of flora which 

" [There is a discrepancy in the Russian text between the common and the Latin name here. The latter 
corresponds to the true firs, whereas the Russian gives the name "spruce. "] 



34 



might have developed in the presence of the mountainous relief that existed 
in the Middle Miocene. Such a zonation might explain some of the instances of 
typically "mixed" Miocene flora. 

Hipparion fauna appeared in the Caucasus in Sarmatian time, having 
migrated from the south along the wide Transcaucasian Isthmus (Figure 16); 
it was probably first discovered in Transcaucasia by Ryabinin (1913). 

Ryabinin found bones and teeth of Hipparion gracile Каир and 
tooth fragments of Rhinocerotidae in the Kyasaman site a;nd on the left bank 
of the lora River. At this locality the vertebrates occur in red clays with 
sandstone intercalations (which also contain algae (Char a cf. escheri) 
and moUusks (Planorbis sp.)) and the same mammalian genera were 
recorded by Ryabinin from gypsiferous clays underlying the Akchagyl beds 
in outcrops in the Katsakhuris-Kedy ridge. 

In 1913, the geologist Dombrovskii, working further south on the right 
bank of the lora River, found in the cliffs of the Eilyar-Ouga and Palan- 
Tikyan ridges a rich locality of Tertiary mammals, known as the El'dar 
locality (1914). 



Kartalinia Plateaus 

El'dar. According to the descriptions of Dombrovskii (1914), and 
Andrianov and Larin (1935), the bone-bearing bed can be traced over 6 km 
from the gorge which connects the El'dar Steppe with the lora River 
floodplain. From west to east the bed grades from a shell conglomerate 
to sandstone, then to limestone, and again to sandstone which grades into 
thick upper Sarmatian clays where the bed eventually peters out. The bones 
(parts of skeletons) occur in pockets, at intervals of 10, 20 and a few 
hundred meters. Ripple marks occur on the top surface of the bone -bearing 
sandstone. 

The lithology of the bed attests to the existence of a low coast and muddy 
bay. The bodies of animals were transported into the bay by weak currents 
and shore waves and were buried in the sandy-clayey bottom sediment. The 
investigators believe that mass mortality in the animals was caused by 
steppe fires, earthquakes and mudflows. 

According to our observations, all the complete bones, including lower 
jaws and ribs, were flattened by the pressure of the overlying beds. Some 
bones were broken in the process of burial; as a consequence, the cavities 
became filled with sand and silt which preserved the original shape of the 
I bone. 

139 The surface of the bones is chestnut or chocolate in color; in fresh 

fracture, the colors are grayish. Permineralization and diagenesis of the 
bone material were not significant. Preserved coprolites were also found. 

Bones abraded by the surf are rare, and in many cases limbs were 
preserved with bones intact. In general, a number of features support the 
conclusions of earlier authors regarding the nature of the burial and the 
mode of preservation. 

Table 2 (Figures 17, 18) lists the species from the Moscow and Baku 
collections identified by Bogachev (1927a), Alekseev (1930), Borisyak and 
Belyaeva (1948). 



35 



TABLE 2. Species and number of mannmal bones from the upper Miocene beds at El'dar 





Number 
ofbones 




Number 
ofbones 


Primates 


1 

4 

6 

I 24 

3 

164 
2 

6 

65 


Artiodactyla 

Sus erymanthius Roth, et Wagn. 

Achtiaria borissiakii Alex 

Ca me lopa rda 1 i s (H e 1 la d о th e r i - 


5 


Carnivora 


. 69 




Tragocerus aff. leskevitschi 




Hyaena of. eximia Gaudry 




Proboscidea 

Mastodon longirostris Каир 

M pentelici Gaudry 


Tragocerus sp. No. 1 

Tragocerus sp. No. 2 


159 
6 


Pinnipedia 








Dinotherium giganteum Каир.... 


5 


Hipparion gracile Каир 


Cetacea 


Uao 






^208 




) 


Aceratherium trans cau casi cum 






Dicerorhinus aff. orien talis 
Schloss 









The El'dar collection may be regarded as a part of the Upper Miocene 
faunal complex of Transcaucasia, since there are no signs of redeposition 
of the bones and most of the species lived at the same time. The most 
abundant species, as far as the collections show, were Hipparion, ibex, 
rhinoceros and giraffe. 

Udabno. This sizable locality containing Hipp ar io n fauna was 
discovered in 1931 by N. A. Gedroits near the village of Udabno, on the 
Kura-Iora water divide of the Garedzhiiskaya Steppe. The locality was later 
excavated by the Museum of Georgia. 

The fossils occur in Sarmatian clays overlain by sandstones of the Shiraki 
formation. This locality and the El'dar locality are similar to each other, 
as are the lists of species from each. The material is in the collections 
of PIN and the Museum of Georgia, and was identified by Burchak- 
Abramovich and Gabashvili (1945, 1950), Tsereteli (1942), and Borisyak 
41 and Belyaeva (1948): 



Primates 



Proboscidea 



Udabnopithecus garedziensis Burtsch. 
et Gab. 



Mastodon sp. 
Dinotherium sp. 



Hyaena sp. 



Hystricidae gen. 



Rodentia 



Perissodactyla 

Hipparion gracile Каир 
Aceratherium sp. 



36 



(40) 




FIGURE 17. Remains of carnivores and hoofed mammals from El'dar 

1, 3- skull of Hyaena sp.; 2 - coprolite of hyena; 4, 5 - tarsal joint and horn axis of T ra g oc eru s 
sp. ; 6, 7 — horn axis and jaw of Gaze 11 a sp. 



37 



Artiodactyla 


A с h t i a ri a sp. 


Sus sp. 


Tragocerus sp. 


Cervidae gen . 


Gaze 11a sp. 



The occurrence of teeth of a large homonid ape (Udabnopithecus 
garedziensis) is of particular interest in connection with studies of 
the origin of man (Figure 19). 




FIGURE 18. Remains of carnivores and hoofed mammals from El'dar 

1 — jaw of С rocu ta eldarica; 2 — incisor of A ce r a theri um t ra n s ca u с a s i cu m; 
3-5 — teeth of Tragocerus leskevitschi (from Bogachev, 1927b) 



Arkneti. This locality is 1.5 km east of the village of Arkneti in 
South Ossetia. It is situated on what was the western part of the 
Transcaucasian bay of the Sarmatian sea. Fragmented bones of mammals 
occur in the yellowish gray loams, 1.5-2.0 m thick. Well-preserved bones, 
some of them joined, and skulls occur in a bone -bearing lense 3 m long 
and 50-70 cm thick. The artiodactyls and perissodactyls have been 
identified by Gabuniya (1952b, 1955c): 



Perissodactyla 
Hipparion sp. (cf. garedzicum n. sp. ) 

Artiodactyla 

Sus sp. 

Eostylocerus sp. 
Dicrocerus salomeae Gab. 



Tragocerus ex gr, leskevitschi Boris. 

Tragocerus sp. 

Gazella cf. gaudryi Schlos. 

Ga ze 11a sp . 

Giraffidae gen. 

Phronetragus arknethensis Gab. 



38 





FIGURE 19. Teeth ofUdabnopithecus garedziensis 

1-4 — Pm^ anterior, outer, inner surfaces (x Ij), upper 
surface (X 2 3); 5-7 —M^ (x Ij) anterior, outer and upper 
surfaces (from Burchak-Abramovich and Gabashvili, 1950) 

i2 The age of the "fauna, " according to Gabuniya, is very close to the age 

of the Udabno "fauna"; it nnay be dated as Upper Sarnaatian (Meotian). 

The existence of large valleys covered with savannah and tropical forest 
vegetation on the piedmont can be inferred from the composition of the 
Sarmatian fauna. Streams originating in the Caucasus l^ange were evidently 
sufficiently powerful to transport the bodies of mastodons and rhinoceroses 
into the coastal bays of the Sarmatian sea. 

As a whole, the Sarmatian faunal complex supports a picture of the 
landscape drawn from the known paleontological data. The plant fossils 
indicate not only the "mixed" character of the Sarmatian flora, but also 
the high degree of its differentiation. 



Stravropol area 

The Hipparion faunal complex occurs in the Sarmatian of the 
Ciscaucasus. The fauna migrated along the coasts and over the plateaus 
and ridges of the central part of the peninsula. Hipparion sp. and 
Dinotherium giganteum Каир are known from the Middle Sarmatian 
limestones near the village of Burlatskoe (Khomenko, 1913a). A small 
giraffe, Camelopardalis parva Weith. , is known from the Upper 
Sarmatian beds near the village of Blagodarnoe in the Stavropol area 
(Pavlova, 1933a). 

Mount Kutsai. A fairly rich locality was discovered by A. A. Ivanchin- 
Pisarev in 1915 near the village of Petrovskoe, northeast of Stavropol. 
Like other Transcaucasian burials of Sarmatian age, it contains both 
marine and terrestrial vertebrates. 



39 



The bone -bearing beds of Mount Kutsai are made of sands and clays 
(Ivanov, 1916). These are underlain by sandy, plastic clays of Middle 
Sarmatian age; soft limey sandstones overlie the fossiliferous sequence. 
The fossils occur in medium-grained marine quartz sands. The material 
is poorly preserved and consists mostly of individual teeth, fragments of 
vertebrae, metapodia and phalanges. Bones of seal and whale occur in the 
lower 2-3 m of the Upper Sarmatian clays and sands. The bones of 
terrestrial forms occur sonnewhat higher, although still within the lower 
3-5 m of the clayey-sandy beds. 

Ivanov's (1916) identification of the terrestrial mammals reveals the 
presence of the Hipparion faunal complex: 



43 



Perissodactyla 

Hipparion gracile Каир. 
Hipparion sp. 
Aceratherium sp. 
Rhinoceros sp. 



Gazella sp. 
Fam. gen. 

Pinnipedia 
Phoca cf. pontica Eichw. 



Artiodactyla 



Listriodon sp. 
S u s sp . 



Cetacea 

Cetotherium priscum Brandt. 
Cetotherium(?) sp. 



Remains of birds, turtles and fishes also occur at this locality. 

Individual occurrences of the bones of rhinoceros (Dicerorhinus 
and Aceratherium) are known from the greenish silts and sands near 
Maikop and from the marine sands near Beshpagir and Stavropol. As a 
whole, the composition of the fauna of Ciscaucasia in the Sarmatian was 
very similar to that of the Transcaucasian fauna. 



In order to identify the origin of the Hipparion faunal complex which 
appeared in the Caucasus, we should compare the Sarmatian "fauna" of the 
Caucasus with other Miocene "faunas" of Eurasia. The following groups are of 
interest: the Miocene "faunas" of Punjab — in the southern foothills of the 
Himalayas in the Salt Range in the Indus River valley, and in the Siwalik 
Range in the Indus and Ganges basins; the Middle Sarmatian "fauna" of the 
Crimea; the Upper Miocene "fauna" of Iran from the eastern shore of Lake 
Urmia; and other occurrences in Asia Minor and the Balkans. 

The "faunas" of the Siwalik and the Salt ranges occur in continental 
deposits up to 7,000 m thick (Colbert, 1935). Over 288 species of mammals 
are known from these Upper Miocene to Lower Pleistocene formations, 
which evolved from the erosion of the newly uplifted mountain range. At 
this time detrital material was deposited on a land surface, sometimes 
partially covered by freshwater bodies. 

The Russian paleontologists and geologists are of the opinion, taken 
probably from Osborn (1921), that the Hipparion "fauna" of southern 
Asia (Siwalik Range) migrated via the Caucasus into Eastern Europe. 

According to Andrusov (1918) the [Hipparion] fauna of the Pikermi 
type appeared in Europe from the east, following two routes: one along 
the northern coast of the gradually shrinking Sarmatian sea and one through 
Asia Minor. 



40 



Andrusov shows that the development of the Hipparion fauna was not 
complete in the Middle Sarmatian but continued through the Upper Sarmiatian, 
reaching its maximum level in the Meotian. In southeastern Europe the 
fauna probably became extinct in the Lower Pontian. 

Borisyak (1928a, p. 376) assumed that the Hipparion fauna migrated 
to Europe from Asia via the Caucasus and the Black Sea landmass. 

According to Bogachev (l938d, pp. 36-37), "the African-Siwalik fauna 
spread through Iran and Transcaucasia onto the growing Main Transcaucasian 
(sic I) Range, and from there, the migration continued across the Stavropol 
Plateau onto the Ukrainian Steppes. The stages of migration are marked 
by fossils which occur along the migration route. " 
44 However, Colbert's lists of the Upper Miocene fauna of Kamlial and 

Lower Siwalik show that there is practically no resemblance to the 
Sarmatian fauna of the Caucasus. 

Nearly 85 species of primates, rodents, carnivores, proboscideans, 
perissodactyls (Hipparion, rhinoceros) and artiodactyls (swine, tragulids, 
deer, giraffe), were noted to 1935 in the upper beds of the Lower Siwalik 
in the Salt area (Chinji). It is significant that no Cavicornia occur in these 
beds. The equids, rhinoceroses without horns, giraffids and primates are 
found in both the Caucasian and the Lower Siwalik faunas. 

The Sevastopol "fauna" in the Crimea, assigned to the Middle Miocene 
age, consists of the following forms (Borisyak and Belyaeva, 1948): 

Carnivora Artiodactyla 

Ictitherium tauricum Boris. Achtiaria expectans Boris, 

Tragocerus leskevitschi Boris, 
Perissodactyla Tragocerus sp. 



Gazella sp, 

Reptilia 



Hipparion gracile var. sebastopolianum 

Boris 
Aceratherium zernovi Boris, 
A. zernovi var, asiaticum Boris. Testudo sp. 

Trionyx sp. 



The Stavropol faunal complex was undoubtedly very similar to the El'dar 
complex, although the zoogeographical relationships between the two are 
not clear. 

Recent geological data indicate that the Hipparion complex could only 
have reached the Crimea from the southwest at the end of the Middle Miocene 
or in the Lower Sarmatian. Later migrations can be ruled out, since the 
Caucasus landmass was probably cut off from the Russian platform and from 
the Crimea by the sea which existed until Pliocene time (Andrusov, 1918, 1926; 
Muratov, 1951). 

The "fauna" occurring in the clays and sands at the southern slope of 
Mount Sahand near Maragheh is of particular interest in the reconstruction 
of the Miocene landscapes and the faunal ties between Asia and the Caucasus. 

According to the identification of Rodler and Weithofer (l890), 
Mecquenem (1924), Bogachev (1928), and Burchak-Abramovich (1952b), 
the following species are represented in the Maragheh "fauna": 



41 



Primates 

Mesopithecus orientalis Kittl 
M. pentelici Gaudry 

Carnivora 

Hyaenarctos maraghanus Mecq, 

Ictitherium hipparionum Gaudry 

I. rob us tu m Gaudry 

Hyaena eximia Wagner 

Meles polaki Kittl 

M. maraghanus Kittl 

Felis brevirostris Croiz. et Job. 

F. a 1 1 ica Wagner 

Machairodus aphanistus Каир 

M. orientalis Kittl 

Tubulidentata 

Oryteropus gaudryi Major 

Proboscidea 
Mastodon pentelici Gaudry 

Perissodactyla 
Aceratherium persia Pohlig 



Rhinoceros morgani Mecq. 
Chalicotherium pentelici Gaudry 
Hipparion mediterraneum Hensel 

Artiodactyla 

Sus (Microstonyx)erymanthius R oth . 
et Wagner 
45 H e 11a d oth er iu m gaudry Mecq. 

Alcicephalus neumayri Rodl. et Weith, 

С a m e lo pa rd a 1 is attica Gaudry 

Urmiabos azerbaldzanicus Burtsch. 

Gazella gaudryi Schlosser 

G. brevicornis Gaudry 

G. capricornis Rodl. et Weith. 

Palaeonyx pallasi Gaudry 

Protoryx carolinae Major 

Antilopinae gen. 1. 

Tragocerus rugosifron Schlos. 

Protragelaphus scozesi Dames. 

Tragelaphus hontom-schindleri Rodl. 

et Weith. 
Helicophora rotundocornis Weith. 
Oiceros rothi Wagner 
O. atropatanes Rodl. et Weith. 
О . bo u 1 e i Mecq. 



Bones of ostrich (Struthio sp.) and Urmiornis maraghanus 
Meq. have also been recorded at the locality. 

The large number of species in the Maragheh "fauna" attests to a high 
degree of universality of the death assemblage. Although the taphonomy 
is not known in detail, it seems likely that the death of the Upper Miocene 
animals can be attributed to a common cause. This might have been toxic 
gases (fumaroles), volcanic ash deposits or mudflows. According to Pohlig 
(1886), the Maragheh "fauna" was buried near the shores of Lake Urmia, 
the water level of which was much higher in the Miocene. The sedimentary 
facies at the locality are similar to the Pikermi locality near Athens and 
the Val d'Arno locality near Florence. The great variability of the 
assemblage is reflected in the occurrence of representatives of different 
ecological habitats: arboreal (e. g., M e s о p i t h e с u s) and savannah and 
steppe (e.g., giraffes, gazelles, ostriches), and of different feeding types: 
carnivores, omnivores, herbivores. A predominance of savannah forms 
over subtropical forest forms indicates that the landscape was of a mixed 
savannah-tropical forest type, like that in the northern parts of the Iranian 
Plateau in the Miocene. 

The Maragheh "fauna" has some "African" elements in it. It is fairly 
close to, though not identical with, the older Sarmatian fauna of 
Transcaucasia. 

According to Stahl (1907) and Oswald (1915-1916), the interior of Iran 
and Anatolia became dry land in the Upper Miocene. Salt and gypsum 
precipitated in the relicts of the Miocene seas. Bogachev and Shishkina 
(1915) discussed the importance of these geological processes in the 
formation of bone -bearing deposits. According to these authors, the climate 
of the country became arid in the Middle Miocene following the uplift and 
folding which subdivided the marine basin into a number of saline lagoons. 



42 



This notion of an arid desert climate is not confirmed by ecological 
analysis of the Maragheh mammals. It is more likely that tropical forests 
existed on the shores of the relict basins, although the climate as a whole 
was dry and hot. 

The Maragheh faunal complex was probably characteristic of all of 
southwest Asia. This is confirmed by geological data on the paleolandforms 
of Asia Minor (Furon, 1955). 
46 Occurrences of Upper Miocene mammals in Asia Minor are known from 

Stambul, upper Gediz, Mugla, Galatia and Cappadocia. The collections 
comprise carnivores (Icht it her ium, Martes, Machairodus), 
proboscideans (mastodon), perissodactyls (H i p p a r i о n and rhinoceros) 
and artiodactyls (giraffe, gazelle, antelope). A similar "fauna" is found on 
Samos Island and in the Balkans, near Athens (Pikermi). The volcanic 
eruptions in Cappadocia, accompanied by ashfalls and mudflows, caused 
death among animals and subsequent burial in layers of tuff. 

As a whole, the Upper Miocene faunas of southwest Asia, the Caucasus, 
the Crimea and the Balkans are similar to one another. However, the order 
of appearance of the H i p p a r i о n faunal complex in each area cannot be 
established until all the material has been studied, a task beyond the 
abilities of one investigator. 



At the end of the Miocene, in Meotian and Pontian times, the seas 
surrounding the Caucasus became shallower and less saline. The Stavropol 
Plateau continued to grow towards the north, the area of the present 
Dagestan Mountains extended to the northeast, the eastern Caucasian gulf 
almost disappeared; and a narrow strait probably existed intermittently 
in the Manych area. 

The major Tertiary uplifts of the Caucasian Range had probably ended 
by that time, and the highland faunal complexes proceeded to form under 
conditions of high-zonal climates and probable local glaciations. 

There is no evidence of any extensive glaciation in the Caucasus in the 
Upper Miocene, as hypothesized by Kovalevskii (1936) in his study of the 
continental formations at Adzhinour. Nor does the paleontological evidence 
support the hypothesis of Kovalevskii and Grossgeim (1936) of extinction 
of the tropical flora and fauna in the Caucasus caused by an Upper Miocene 
glaciation. Neither is there any evidence that the Giinz and Mindel glaciers 
were so large that they could "plough" and fill the Alazan-Agrichai valley 
in eastern Transcaucasia, which, in any case, is probably not that old. 

Grossgeim (1936, 1948) believed that in Meotian time the southern 
xerophilous flora invaded the Caucasus. However, the fossil record shows 
that the only flora known is the mesophilous forest flora in Gurie [Western 
Georgia] in which deciduous species (Fagus orientalis, Acer 
trilobatum) predominated, but which also included evergreen subtropical 
forms (Rhododendron ponticum) and some admixture of conifers 
(Taxus grandis). Mesophilous plants were also recorded at Nakhichevan 
in the Lower Pliocene rock salt deposit (with paper shales): alder, mazzard, 
hop, hornbeam, willow, sedge and reed (Palikin 1936, Bogachev and 
Shishkina 1915). 

Fossil plants and animals of the highlands are not known from that 
period. However, it is possible to assume that, in addition to some endemic 
evolution, cold -climate flora and fauna migrated to the Caucasus over the 
ranges of the Alpine region. 

43 



47 The extant forms which evolved in the Upper Miocene are the Asia Minor 

hamsters (M e s о с r i с e t u s), Prometheomys s с h ap о s с h n i к о v i 
and the wild Caucasian goat. 

The Uppermost Miocene terrestrial fauna which inhabited the piedmont 
plains is as yet little known and poorly dated. The Hipparion faunal 
complex continued its existence on the plains around the Caucasus mountain 
system, particularly on those to the southeast. 

Rare occurrences of mammals are known in eastern Transcaucasia and 
in western Ciscaucasia. One was discovered by N. A. Kudryavtsev in 1935 
on a northern spur of the Kakhetian ridge in Transcaucasia. 

The stratigraphic position of that locality is not clear, although it has 
been tentatively dated as Upper Miocene or Lower Pliocene. According to 
Orlov (1936b) the bone-bearing bed is exposed on the left slope of a deep 
gorge which cuts through the village of Dzhaparidze (near Tsiteli-Tskharo) 
toward the Alazan River. Scattered and broken limbbones, jaws and teeth 
of mammals occur in marine (?) clays overlain by continental clays 
alternating with layers of fossil soils. The following species are known 
from the locality: 

Proboscidea Artiodactyla 

Fam. gen, Sus sp. 

Cervidae gen. 

Perissodactyla Giraffidae gen, 

G a zella sp. 
Hippanon gracile Каир 

Rhinocerotidae gen. 

In Ciscaucasia the Upper Miocene mammals were collected by 
Ya.M. Eglon in 1940 from the Meotian (?) clays on the right bank of the 
Kuban River near Armavir. Individual fragments of permineralized bones 
were identified as Hipparion sp. , antelope about the size of saiga 
(Antilopinae gen.), small beaver (Castoridae gen.), small seal (Phoca 
sp. ) and saber-tooth (Machairodus sp. ). 

Fragments of molars of a rodent (Muridae), about the size of a house 
mouse, were found by V. Sizov in the Miocene beds of the River Aksai 
ravine in the Grozny Region at the boundary of the Upper Sarmatian and 
Meotian yellow -green clays. Remains of grasses, freshwater ostracods 
and otoliths of gobies (Gobius) were also collected. 

The vertebrates from the continental deposits near Novocherkassk, in 
the south of the Russian Plain, were tentatively dated Meotian age. 

The bodies of animals were probably carried by the paleo-Don into the 
Tanais Gulf of the Meotian sea where they were buried in the crossbedded 
white sands. The stratigraphic position of the sands is "between the Pontian 
and the eroded Sarmatian" (Zhizhchenko, Kolesnikov, and Eberzin, 1940). 
The sands are up to 12 m thick. Very thin intercalations of bluish gray clay 
with diatoms, silicified wood and teeth of proboscideans occur in the sands. 

Near the village of Yanov, mollusks ( Congeria and N e r i t i n a ) 
and perch bones (Perca sp. ) were collected from the sands. The mammals 
collected by Lisitsyn and Bykodorov near the villages of Yanov and Popovka 
and in the quarries along Tuzlovka River were identified by Khomenko. 
Bogachev and Sokolov (1954) as belonging to four species: 



44 



48 Proboscidea Dinotherium gigantheum Каир 

Mastodon borsoni Hays. Artiodactyla 

M. cf. tapiroides Cuv, Palaeoryx Ion g i ce ph a lus Soc. 

Thus the end of the Miocene in the Caucasus was characterized by the 
growth of landmasses, the appearance of northern species in the flora and 
the development of altitudinal zones of vegetation resulting from cooling of 
the climate and nnountain -building movennents .These processes, taken together, 
promoted the rapid evolution of horses, Cavicornia, ruminants and deer, 
and a concomitant decline of giraffids. 



PLIOCENE 

In the Caspian region, four stratigraphic subdivisions of the Pliocene 
can be recognized through the fauna: Pontian, Balakhany, Akchagyl and 
Apsheron. In the Black Sea region, there are five subdivisions in the same 
period: the Pontian and four subdivisions above it corresponding to the three 
upper subdivisions of the Caspian: Cimmierian, Kuyal'nitsk, Gurie and 
Chauda. 

The outline of the Caucasian Peninsula did not change essentially from 
the Upper Miocene to the Lower Pliocene (Lower Pontian). The peninsula 
extended into a slightly saline lake-sea, not connected with the ocean 
(Andrusov, 1918). Two large open bays cut into the northern coast of the 
lake: the Kuban bay in the northwest and the Terek bay in the northeast . 
In the western part of present-day Colchis, the Rion bay extended to the 
longitude of Kutaisi and to the east lay the Samur and Kura bays. The thick 
Pontian sediments and the littoral conglomerates (Kolesnikov, Zhizhchenko, 
and Eberzin, 1940, p. 402, map) indicate that all of these but the Kuban were 
very deep. Their depth suggests a considerable uplift in the coastal areas 
of the region which today comprises Abkhazia and northern Azerbaidzhan. 

The broad land connections with southwest Asia allowed the nnigration 
of southern plants and animals to the Caucasus. 

A number of geologists (for example, Sokolov, 1904) have noted that 
in Lower Pontian time the climate became cooler and the rivers of the 
Russian Plain periodically froze. 

Nevertheless, the climate of the Caucasus remained fairly warm in 
Lower Pliocene time. The abundant plant reniains in the Yergeni beds (50 km 
north of Stalingrad, in the Ilovlya River Basin) represent deciduous, warm- 
climate flora of the Upper Miocene-Lower Pliocene. The flora consisted 
of Corilus fossilis, Alnus incana, Quercus sp., Castanea 
sp. , Parrotia persica, and A r a 1 i a с e a (identified by Baranov, 1952). 

The occurrence of the Russian pea shrub (Parrotia persica) in the 
Yergeni beds is a good indicator of a warmi climate in the Lower Pliocene, 
since this species only survived through the Pleistocene south of the 
Caspian sea, i.e. , 10° in latitude farther south. 

With the Middle Pontian, the climate in the Caucasus became warmer, 
as evidenced by plants collected in stratigraphic sections in the Pontian 
sediments at Cape Pitsunda (Mchedlishvili, 1954b). 



45 



At the end of Pontian time, the seas receded from Ciscaucasia and the 
water remained in the Caspian region only in the southern part of the basin 
(Figure 20). Extensive semideserts probably formed in eastern Ciscaucasia 
and eastern Transcaucasia. 
49 Later, in Cimmerian time, the climate became almost tropical, as 

indicated by the occurrence of banded iron-ores in the Taman and Kerch 
peninsulas and in the foothills of the northern Caucasus. Tropical conditions 
can also be inferred from the occurrence of tropical plants in marine 
littoral sediments (Mchedlishvili, (1954a) and in volcanic tuffs at Goderdzi 
on the Adzhar-Imeretia ridge in Transcaucasia (Palibin, 1936). The plant 
remains include Ficus, Araucaria and palms. 




FIGURE 20. Cimmerian and Balakhany basins (from Kolesnikov; see Zhizhchenko and others, 1940) 
Dots indicate main localities of land vertebrates 



The land connection between the Caucasian Peninsula and the Russian 
platform was established for the first time in the Cimmerian; this allowed 
migration of plant and animal species in both directions. 

At that time the fauna of the deserts of central Asia could have migrated 
far west to the eastern plains of the Caucasian Isthmus. During the 
Cimmerian, the configurations of the Black and Azov seas were similar 
to their Recent configurations. 

Lx)wer and Middle Pliocene terrestrial vertebrates of the Caucasus occur 
primarily in continental deposits. 



46 



Stavropol Plateau 

The largest and most complete body of Pliocene material was found 
10 km west of Stavropol in a locality of sandy sediments which were exposed 
50 in the Kosyakin quarry. A bone-bearing lens, 7-9 m thick and 90 m 

wide, was formed in a channel of a Pliocene river which cut through the 
Sarmatian limestones at the edge of the Stavropol Plateau (Kaspiev, 1939; 
Gnilovskoi and Egorov, 1955). 

Seven beds can be distinguished in the old alluvium. The lithology of the 
beds varies from fine-grained, grayish white sand to gravels with clayey 
intercalations and clayey pebbles up to 1.5 cm in diameter (Figures 21, 22). 




FIGURE 21, Stratigraphic section in the Kosyakin quarry 

a, b — loams with humus; c-g — bone-bearing river sands, some of them 
crossbedded. Numbers on the right indicate thickness in meters 



Our observations show that the pebbles in the lower part of the section 
are made of Sarmatian local rocks, rounded bones of Sarmatian marine 
mammals and, more often. Pliocene land mammals. 

Bones of animals, fractured and redeposited many times, were retained 
in the gravelly sands which were deposited at [current] velocities of 
1.3 m/sec (lower bed) and 0.1 m/sec (upper bed). 

The bones were preserved in various stages of abrasion, from complete 
specimens to rounded pebbles. 

Up to 92% of the bones collected in the quarry are free of organic 
matter and nearly free of permineralization. The bones resemble friable 
marl or chalk. The surface color of the bones is cream or grayish. All the 
cracks in the bones of this type are iron-stained. Manganese and iron 
dendrites occur rarely in the bones. Thin wavy grooves, the marks of 



47 



51 



Recent plant roots, are common on the bones. Notches of irregular shape, 
fornned by water erosion and aquatic invertebrates at the time of deposition, 
were observed on some bones. 

The bones in this group are, for the most part, intact; fractures are 
mainly the result of careless collecting during quarrying. The remaining 
8% of the bones are gray, and numerous dendrites give the surfaces a stained 
appearance. As a rule, these bones are more strongly permineralized and 
considerably heavier than the "chalky" bones. The tooth enamel is always 
well preserved; its color is either natural or light pea green or light pink. 

Tooth impressions of carnivores are rare: only nine bones show signs 
of gnawing on the epiphyses. 




FIGURE 22. Kosyakin quarry 



Photograph by author, 1952 



The occasional vertebrae of Sarmatian whales and seals, derived from 
the Sarmatian limestones, are easily distinguishable from Pliocene bones 
by the heavy permineralization and grayish brown color. 

Variations in the conditions of deposition and in the exposure times on 
river banks and spits during redeposition account for the differences in the 
state of preservation of the Pliocene bones. 

It seems likely that the physical environment of the valley and stream 
of the Stavropol paleoriver was similar to the middle parts of the Don, the 
Volga and the Ural in Quaternary time. Bodies of dead animals accumulated 
in the oxbow lakes, as well as in the mainstream and backwater. 

In the process of erosion of older marine sediments, the river carried 
the bone material over short distances and redeposited it. 

The direction of the stream flow has not yet been established. Perhaps 
when the sea level was low and the Stavropol area slightly elevated, the 



48 



river source was somewhere on the Russian Plain (e.g., near Yergeni) 
rather than on the Caucasus. 

In spite of the different ages of the bones at the locality, most of the 
Pliocene species can undoubtedly be regarded as one faunal complex. 

The fragmented material makes identification difficult, and most of the 
species have not yet been described in detail. 

A preliminary list of the species is given in Table 3 (collections of PIN 
and ZIN: identification by Belyaeva (1940b, 1944), Borisyak and Belyaeva 
(1948), Argiropulo (1939b, 1940c), and Vereshchagin (1954)). 

TABLE 3. Composition of the Pliocene fauna and number of bones from the Kosyakin quarry 



Number 
of bones 



Number 
of bones 



52 



Insectivora 

Crocidura sp 

T a Ipa sp 

Desman a sp 

Carnivora 

Ursus cf, arvernensis Croiz, 
Dinoc;'-)n cf. thenardi Jourdan 

Canidae gen 

Felis cf, issiod orensis Croiz. 
et Job 

Lagomorpha 

Lepus sp , 

Ochotona cf. antiqua , 

Rodentia 

Amblycastor caucasicus Arg. 

Steneofiber sp 

Cricetus sp 

Mus sp , 



12 
4 
3 



25 

2 



1 

3 

37 

1 



Proboscidea 

Anancus arvernensis Croiz, etjob, 
Dinotherium sp 

Perissodactyla 

Hipparion sp 

Hipparion gracile Каир 

Tapirus cf, arvernensis Dev. et 

Bouill 

Dicerorhinus orientalis Schlos. 
Aceratherium cf, incisivum Каир 
Rhinccerotidae gen, (cf, Chilo- 

therium) 

Rhinocerotidae gen 

Artiodactyla 

P гор о t a mo ch о e r u s provincialis 

Gerv 

Procapreolus sp 

Pliocervus sp 

Pseudalces sp 

Giraffidae gen. (cf. Sivatherium) . . . 
Gazella sp 



21 
12 



3 
72 



1 
33 



10 
7 
9 
2 
2 
5 



Note. Asterisk indicates author's collection of 1952. 



53 



In addition to the land mammals given in Table 3, the material includes 
birds (Charadrius cf. morinellus L. ), aquatic turtles 
(Trionychiidae — 14 carapace fragments), the lower jaw of a small lizard 
(Lacertilia), and bones of small anurans (frogs and toads) (Figures 23-25). 

The faunal complex of the ancient Stavropol area was well developed. 
The Stavropol mastodons, tapirs, rhinoceroses and warthogs lived among 
subtropical mesophilous plants. Groves of tugai vegetation near the rivers 
probably alternated with open meadows inhabited by moles, hamsters 
and pikas . 

In the preceding sections we reached certain taphonomic conclusions 
and outlined a landscape scheme. These judgments are, to an extent. 



49 



confirmed by the following phenomena: the predominance of fossils of 
animals associated with river valley thickets (rhinoceros, swine; Figures 
26 and 27), the occurrence of beaver, tapir and aquatic turtle fossils in 
swamps, oxbow lakes and channels, and the iron stain which appears on 
these fossils. The presence of small desmans in the Pliocene river indicates 




FIGURE 23. Bones of smaller vertebrates from the Kosyakin quarry 

1 — jaw of lizard (Lacertidae) ; 2 — humerus of T a Ip a sp. ; 3 — jaw of 
Desman a sp, ; 4, 5 — humerus and calcaneus of Lep u s sp, ; 6, 7 — upper 
molars and femur of St e ne of iber sp. ; 8 — jaw of С r icet us sp. 



the antiquity of the erosion valley and the stability of the hydrological regime 
of the rivers, which is similar in this respect to the regime of the Don and 
the Volga. 

The Stavropol complex is not of Upper Pliocene age; Borisyak (1943) 
has dated it Pontian. The Lower Pliocene age is indicated by the absence 
of elephants and horses proper. According to Argiropulo (1932, 1940c), 
the Caucasian Amblycastor is very close to the Upper Miocene species 
of this genus occurring in Mongolia and North America. The deer in the 
Stavropol faunal complex are of the Lower and Middle Pliocene types. The 
56 Stavropol faunal complex can be tentatively dated as Lower Pliocene on the 
bases of accepted geological data and the composition of the complex. 



50 



(54) 




FIGURE 24. Remains of carnivores and proboscideans from the Kosyakin quarry 

1 3-jawsofUrsus of. arvernensis; 2 - jaw of D in осу on cf. thenardi (xi); 4-premolar 

of Canidae gen. (X2); 5, 6 - jaw and shoulder of F el is cf. i s s io do re ns i s; 7-M4Anancus 

arvernensis 



51 



(55) 




FIGURE 25. Remains of hoofed mammals from the Kosyakin quarry 

1, 2 — metapodia of H ipparion gracile and Hipparion sp. ; 3 — jaw of T a p i rus cf. 
arvernensis; 4 — jaw of Propota mochoerus prov in cia lis; 5, 6 — horn and jaw of P roca • 
preolus sp.; 7 — jaw of Cervidae gen. 



52 



It is younger than either the Taraklian connplex of Moldavia (Khomenko, 
1914b) or the Pavlodar complex on the Irtysh River and is quite close in 
age to the Malusteni complex in Rumania (Simionescu, 1930). 



Some geologists correlate the continental elastics, in the foothills of the 
Ciscaucasus, 700-800 m thick, with the Cimmerian marine sediments. The 
continental elastics in Kabarda and North Ossetia have been described by 




FIGURE 26. Jaw of D icerorhinus ori en talis from the Kosyakin quarry 



Shvetsov (1928), as follows: the continental sequence resembles a moraine 
and consists of loams, sands and tuff breccias with large boulders of erupted 
material in some places. The continental beds form a scarp on the northern 
slopes of the Chernye-Gory Mountains; in the Zmeika ridge the continental 
beds overlie the Sarmatian clays. The beds were probably formed by 
mudflows carrying great volumes of poorly sorted material from the 
(,„ mountain slopes. The fossils in the continental beds are land and freshwater 
mollusks (Helix sp. , Paludina sp. , Neritina sp.), grasses and 
leaves, andbonesof land vertebrates. M. V. Pavlova has identified an 
antler (Cervus (?)) and a "rat jaw" in Shvetsov's collection. 

Sands and clays, 1,000-1,400 m thick, accumulated in eastern 
Transcaucasia in Cimmerian time; those on the Apsheron Peninsula 
comprise a large oil reservoir, known as the Balakhany productive beds. 

There are various opinions on the history of the productive beds. 
Kolesnikov, in a review of the theories of their origin (Zhizhchenko, 
Kolesnikov and Eberzin, 1940), concludes that the beds were formed in a 
"developing tectonic basin, " which is to say that they were formed from the 
clastic products of the erosion of the eastern Caucasian Mountains which 
accumulated in subsiding shallow -water basins. Land vertebrates are very 
rare in the productive beds. 

In 1932 V. Podgornova discovered fossil mammals in light-colored, 
fine-grained sandstones in the productive beds to the east of the Lok-Batan 
volcano south of Baku. According to Bogachev's (l938b, 1941) and our 
identifications, the following species occur at the locality: 



53 



Camivora 
Vulpes khomenkoi Bog . 

Artiodactyla 

Cervus (cf. Eu cla tJ ocerus)sp. 
G a z e 1 1 a sp . 



58 



The sheatfish (Siluris glanis L.) has also been recorded at the 
locality. 

Two complete sets of upper molars and jaw fragments of fox have been 
found embedded in the light gray, fine-grained sandstone. The bones were 
considerably permineralized; the replacing substance is straw-colored or 
pink in fresh fracture. The tooth enamel is black and glossy. The material 
includes a fragment of gazelle skull with complete sets of upper molars and 
two complete metacarpals of deer. There is no basis for identifying, as 
Bogachev did, the Middle Pliocene deer from Lok-Batan with the 
Quaternary red deer and goitered gazelle. The bones at the locality were 
probably deposited in the delta of one of the rivers draining the eastern 
slopes of the Caucasus, a river similar to the present-day Sumgait and 
Pirsagat rivers. 

The layers adjoining the sandstones of the productive beds contain 
freshwater mollusks (Golubyatnikov, 1925): Planorbis cornu var. 
manteli Dunk., P. costatus Klein, Limnaea armanensis Noul. 
and other limnaeids, Unio jasamalicus D. Golub. and Melania 
glacilicosta Sandb. The alga, С h a r a sp. , also occurs in these beds. 
It is doubtful that such a collection of species could exist under desert 
conditions in rivers and lakes of variable salinity. The leguminous plant, 
Albizzioxylon hyrcanum, which is characteristic of subtropical 
conditions, is known from the Balakhany beds of the Shiraki Steppe 
[Leninakan Steppe], west of the Apsheron locality. 

The fossils occurring in the Balakhany beds indicate that the differentiation 
of faunal complexes followed the differentiation of the landscapes in the 
Caucasus from subtropical forests to hot dry semideserts. 



(56) 




I < i ■ ii i I 



FIGURE 27. Canine (?) of R hinocerotidae gen. from the Kosyakin quarry 



54 



The Middle Pliocene mammals of the southwestern part of the Russian 
Plain evolved from eastern Mediterranean Miocene faunas. The Middle 
Pliocene assemblages of mammals in Moldavia and in the Ukraine are 
particularly important to an understanding of the evolution of the Ciscaucasian 
fauna. The Moldavian assemblage occurs in gravelly-sandy fanglomerates 
in the basins of the Salcia, Kagul and Prut rivers; the Ukrainian 
assemblage is known from limestone catacombs, filled with red clay, in 
Odessa. 

The Middle Pliocene "fauna" of Moldavia consists of ape, rodent, carnivore, 
rhinoceros. Hi pp ar i о n , hippopotamus, camel, deer and antelope. 
Lagomorph and rodent are represented by the Recent genera of hares, pikas, 
squirrels, beavers, porcupines and mole rats (Khomenko, 1914a, 1915; 
Borisyak, 1943). 




FIGURE 28. Kuyal'nitsk and Akchagyl basins (from Kolesnikov; see Zhizhchenko and others, 1940) 
Dots indicate main localities 



The "fauna" of the catacombs in Odessa consists of nearly 30 species of 
insectivore, rodent, carnivore, proboscidean, artiodactyl and an ostrich 
(Gritsai, 1938, 1939; Borisyak, 1943; Pidoplichko, 1954). 

The faunal complexes of the areas north of the Black Sea reflect the 
variability of the landforms: wooded creek and river valleys alternating 
with steppes. Some of these complexes might have extended southeastward 
to Ciscaucasia. 

Both the Moldavia (Roussillonian) and the Odessa Middle Pliocene 
"faunas" contain a number of subtropical forms. 

The occurrences of ape and hippopotamus in the Roussillonian of the 
eastern Mediterranean area are evidence of the warming of the climate in 
post-Pontian time. The climate of the north edge of the Cimmerian basin 



55 



at the end of the Middle Pliocene was probably similar to the Recent climate 
of southern Italy, Spain and France, while in Ciscaucasia the climate was 
probably of a more continental t^'pe. 

During the Upper Pliocene, the configuration of the Caucasian Isthmus 
was similar to its present form, but it was intermittently cut off in the north 
by narrow straits in the Manych region, which temporarily severed the land 
connection between the Caspian Sea and the Black Sea basins. 



In the Black Sea region most of the Upper Pliocene localities are situated 
along the sea coast. Fossils can probably be found in the zone of contact 
between deltaic deposits and gravelly fanglomerates of the paleo-Don, paleo- 
Donets, paleo-Dnieper and other ancient streams of the Kuyal'nitsk and 
post-Kuyal'nitsk basin (Figure 28). 

59 A number of authors in the last and present centuries (Fischer von 
Waldheim (1809), Eichwald (1850), Xordman (1858-60), Pavlova (1895), 
Bogachev (1923-24, 1938d), Sherstyukov (1926, 1927) and Grigor (1929)), 
have noted the following mammals from the Upper Pliocene of the Azov 
Sea and Kuban River regions: 

Carnivora E. meridionalis Nesti 

E.antiquus Falc. 
Hyaena sp. 

Meles sp. 

Perissodactyla 

Rodentia Equus stenonis Cocchi 

Trogontherium cuvieri Fisch. 

Artiodactyla 

Proboscidea Camelus sp. 

С e г V u s SD 
Mastodon arvernensis Croiz. etjob. 

Elephas lyrodon Weithofer ^ "' 

Upper Pliocene mammals in the Azov Sea region were collected by 
Gromov (1933, 1936) in 1933-1936 in the gravelly sands near the villages 
of Morskaya, Merzhanovka and Khapry, and in the Volovaya gulley near 
Taganrog. Similar fossils were collected by the author from the Leventsovka 
quarries near Rostov in 1954. Fragmentary iron-stained and well-rounded 
bones from the Khapry sand indicate that the bones were buried in alluvial 
fans and in sediments exposed to marine abrasion. Some of the bones are 
marked by bore holes made by marine inv^ertebrates, showing that they had 
been deposited in the sea (Figure 29). 

60 According to the identifications of Gromov (1948), Borisyak and Belyaeva 
(1948), and the author, the following species occur in the Azov Sea region: 



Proboscidea 



Carnivora 



Mastodon sp. 

Canis cf. lupus L. Elephas planifrons (?)Falc. 

Canis sp. E, meridionalis Nesti 

Ursus sp. 

Hyaena sp. Perissodactyla 

Machairodus sp. 

Hipparion sp. 

„, . Equusstenonis Nesti (cf. major Boule) 

Rodentia 

Rhinoceros sp. 

Trogontherium cuvieri Fisch. E lasmo theriu m sp. 



56 



Artiodactyla 

Sussp. Eucladoceruspliotarandoides A lies. 

Camelus (P a г a с a m elu s) cf. ku j a 1 n i к e n s i s Cervus cf. elaphus L. 

Chom. Cervus sp. 

Fragments of ostrich long bones have also been collected in the region 
(S t r u t h i о sp. ). 




FIGURE 29. Jaw of Elephas meridionalis with borings caused by invertebrates 
From the Khapry sands on the right bank of the Don River (Orig. ) 

Although the species listed are of different ages, as pointed out by 
Bogachev (1923-1924), they may, nevertheless, be assumed to represent 
one faunal complex. Since the same species also occur farther west, the 
complex may be considered characteristic of the Upper Pliocene of the 
southern part of the Russian Plain. 



Trans-Kuban Plain 



61 



The burials in the gravels of the Trans -Kuban Plain along the 
Psekups River are in the main similar to the northern Azov Sea 
localities. The material was collected in 1930-1932 by Gromov (1935b, 
1937, 1948) near the villages of Bakinskaya and Saratovskaya on the left 
bank of the Psekups River. Two floodplain terraces can be distinguished 
on the bank: the lower terrace of Wiirm age and the upper terrace of Riss age, 

Gromov (1948) has demonstrated the presence of freshwater basins in 
the foothills of western Caucasia in geological sections which show a 
variation from fast -flowing stream deposits to swamp deposits. 

According to Gromov (1948, p. 56) the following species occur in the 
lower and middle parts of the sedimentary sequence: 



Proboscidea 
Elephas meridionalis Nesti 



— teeth and limb bones 



57 



Perissodactyla 

Rhinoceros etruscus Falc. 
Equus stenonis Cocchi 
Eq.ius sp. (aff. stenonis) 

Artiodactyla 

Bison sp. 

Bos sp. (Leptobos) 

Eucladocerus pliotarandoides A less. 

С ervus sp. (Rusa) 

Cervus sp. (ex. gr. polycladus?) 

С e r vus sp. 



— nearly complete skull and skeletal components 

— metapodia, phalanges 

— limb bones 



— broken skull 

— broken skull 

— skull with a horn and horn fragments 

— fragments of horns 

— fragments of limb bones of a very large deer 
(larger than elk! ) 



62 



The material also included the lower part of an ostrich tibia (Strut i о 
sp. ). 

The above-mentioned material from the Psekups was lost during the 
war before it could be described. 

The material preserved in PIN includes an antler of a deer (Eucla- 
docerus, Figure 30) and a skull of the Etruscan rhinoceros. The material 
is straw-colored and strongly impregnated with iron. 

From the presence in the Trans -Kuban Plain of intact skulls and horns, 
it can be inferred that during the Upper Pliocene quiet rivers flowed over 
this area into large bodies of still water. Such landforms could exist only 
at the stage of more advanced peneplanation of western Caucasia. 

East of the Psekups there have been individual occurrences of large 
Upper Pliocene mammals in the gravel beds deposited in piedmont basins, 
such as the bones and teeth of elephant (Elephas cf. planifrons Falc. 
and E. meridionalis Nesti) observed along the Laba and Kuban rivers, 
in the region of Tshchikskie plavni-'= (Navozova, 1951), near the villages of 
Novo- Labinskaya, Nekrasovskaya, and Grigoropolisskaya (Gromov, 1937), 
and south of Armavir. The occurrences of southern elephant, rhinoceros 
(Rhinoceros cf. etruscus Jaeg. ) and deer (Eucladocerus 
pliotarandoides Aless . ) are more frequent in the ferruginous sand 
near the village of Voskresenskaya, west of Stavropol, and in the lower part 
of the river sands in the Girei quarry near the Kropotkin station (Figure 3l). 

At the time when this faunal complex lived in Ciscaucasia the climate 
was mild-temperate; broadleaf forests of the Colchis and Hyrcania type 
covered considerable areas of western Ciscaucasia. Bos, horse and 
ostrich fossils evidence a progressive development of savannah-type 
grasslands. The animals inhabiting the piedmont plains were probably 
often killed by torrential streams and buried in masses of deposited gravel. 

In other cases, accumulation of bones in the gravels was caused by the 
erosion of bone -bearing lenses by later streams. Therefore, the Psekups 
and the Taman faunal complexes (see below) probably antedated the 
formation of the bone-bearing gravels in the Ciscaucasian plains. 

Most of the fossiliferous strata in the Psekups, Laba and Kuban areas 
are usually correlated with the Akchagyl and sometimes with the Apsheron 
stage of marine sedimentation in the Caspian region. 



* [Periodically flooded areas with Phragmites, Тур ha and С a re x in river deltas and bottomlands in the 
southern part of the U.S. S. R. ] 



58 




'■''■'■■''' 



FIGURE 30, Horn of Eucladocerus p lio ta га nd о i des 
from the gravels on the Psekups 



Taman Peninsula 

The bone -bearing formations of the Taman Peninsula are slightly younger 
than the lower beds in the Psekups area. The fossiliferous localities on the 
Taman Peninsula are the Sinyaya and Kapustina gulleys and Sennaya, Fonta- 
lovskaya and Kuchugury (Figure 32). The bone-bearing breccia at Sinyaya 
gulley was discovered by Gubkin (1914) in 1912. He identified the freshwater 
sands exposed on the northern coast of the peninsula as of Basal Pleistocene. 
The sands contain numerous freshwater mollusks of the genera Vivipara, 
Bythinia, Dreissensia, Unio, and other forms (Figure 33). Broken, 
sometimes rounded, bones were found to be heavily permineralized, 
particularly at the Sennaya and Fontalovskaya localities. Permineralized 
excreta of hyenas, swine and ruminants also occur in the sandy-gravelly 
beds at Sennaya. The material from the localities mentioned is housed in 



59 



33 



the ZIN and PIN collections, and has been identified by Borisyak (1914), 
Belyaeva (1925, 1933a, b), Vereshchagin (1951a, 1957a) and Burchak- 
Abramovich (1952a). The species identified are given in Table 4 and 
Figure 34. 

A vertebra of a large fish was taken from the ferruginous sand near 
Sennaya. A seal phalange, resembling those of the Caspian seal (Phoca 
с as pi с a Gmel. ) (Kirpichnikov, 1953) is probably not from the same beds 
from which land vertebrates were taken at Sinyaya gulley. 

A taphonomical and ecological analysis of the species studied suggests 
that they are part of one faunal complex. The fauna lived on the land which 
formed between Ciscaucasia and the Danube River during one of the 
regressions in the post-Kuyal'nitsk (Apsheron) basin. 

The presence of horse and bison (species adapted to steppe life), of rhinocer- 
os, forest antelope, deer and elephant (species adapted to forest life), and of 
beaver (a species adapted to lake and river life) indicates that western Cis- 
caucasia was a plain with dense vegetation alongthe rivers. The climate was 
probably moderately warm. From the morphogenetic point of view, this faunal 
complex is very close to the Villafranchian complex of Western Europe and 
the Nehavend complex of southwestern Asia. However, a number of 




FIGURE 31, Chauda and Apsheron basins (from Kolesnikov; see Zhizhchenko and others, 1940) 
Dots indicate main localities 

Taman species are indistinguishable from the species of the Khapry complex 
(beaver Trogontherium, southern elephant, camel) (Vereshchagin, 
64 1957a). Similar species also occur in the Apsheron beds of eastern 
Ciscaucasia and Transcaucasia. 



The evolution of Upper Pliocene fauna complexes in the Caspian region 
was studied at localities situated along the margins of the Akchagyl and 
Apsheron basins of the Caspian depression. 

In Akchagyl time the axis of subsidence in the northern Caspian became 
oriented in a north-south direction, as opposed to its earlier east-west 



1704 



60 



f со) TABLE 4. Upper Pliocene mammals and number of bones from five localities on the Taman Peninsula 



Number 
of bones 



Number 
of bones 



Carnivora 

Canis tamanensis N, Ver 

Panthera sp 

Rodentia 

Castor tamanensis N, Ver 

Trogontherium cuvieri Fisch. . . . 

Proboscidea 

Elephas meridionalis Nesti" 

E. trogontherii Pohl 

E. antiquus Falc 

Perissodactyla 
Elasmotherium caucasicum Boris. 
"• According to Garutt (1958) this is Phan 



803 



66 



Rhinoceros cf. etruscus Fair. . . 
Equus aff. s ussen bo rn en s is Wiist . 

Artiodactyla 

Sus tamanensis N. Ver 

Camelus cf. ku j a In i kensis Chom. 
Eucladocerus sp. 
Megaceros cf. euryceros Aldr. 
Tamanalces caucasicus N. Ver. 

Cervidae gen 

G a zella sp 

Tragelaphus sp 

Strepsicerotini gen. (cf. Taurotragus) 
Bison cf. schoetensacki Freud. . . 
Bison sp . 



5 
54 



2 
1 

22 
1 
3 

41 
2 
2 

18 

12 
2 



agorolcxodon m a m m о n t о i d es. 



te; .: : ■ ■ - •; ■■.--■, ^ЯИ^^^^^^^^^^^^^^^^^^^^^^^^^^ИИ^^^^^^^И 


^^ 




НИММм»» «3W* ^.^ _^ 


.#9^****^- 





FIGURE 32. Cliffs and landslides on the northern coast of the Taman Peninsula, near Cape Litvinov 

Photograph by author, 1954 



61 



(65) 




E Я 



<^- E 



СЛ 


^ 


с 

.2£ 




■i 




CO 




r 


CO 


Ё 


*-' 


Ul 


(U 


ai 


о 


<fl 


D 
О 


1 


-o 



62 



65 



orientation. Akchagyl marine sediments can be traced from eastern 
Transcaucasia and the foothills of the Kopet-Dag in the southeast to Kazan 
and Ufa in the north (Arkhangel'skii, 1934; Kovalevskii, 1933; Kolesnikov, 
(Zhizhchenko, Kolesnikov and Eberzin, 1940)). In Akchagyl time, the rivers 
draining the Caucasus Range entered the narrow Kura bay which extended 
almost to Tiflis, and the wide Terek bay which extended to Mozdok. The 
rivers transported the bodies of dead animals to the plains and to the 
coastal bays. 

During the transgressive naaximum, the Apsheron and the Kuyal'nitsk 
basins were connected by a narrow strait south of the Stavropol Plateau 
(Gatuev, 1932). 

This growth of the relict marine basin at the time of deposition of the 
productive beds has been explained by Kovalevskii (1933) as the result of 
marine transgression into the Caspian region. However, his hypothesis 
on connections with the boreal sea is not supported by recent data on 
molluscan faunas which occur in the northern part of the basin. 

Gradual climatic changes probably persisted throughout Akchagyl time. 
However, the glaciation of the Greater Caucasus, assumed by Kovalevskii 
66 (1936), had not yet begun. A reoccurrence of volcanic activity in the Akchagyl is 
indicated by thin layers of ash and punnice which appear in the massive 
loams and gravels of the foothills region. The Akchagyl basin, which 
covered a great area of the Russian Plain, was a natural barrier on the 
migration routes between central Asia and Ciscaucasia (Figure 28). 

The Akchagyl land flora is known from the shores of the Kura bay, near 
Naftalan, from Kakhetia and from the Shiraki Steppe. The following species 
have been identified by Palibin (1936): oriential beech (Fagus 
orientalis), oak (Que reus sp. ), willow (Salix alba), common 
pomegranate (Punica granatum), elm (Zelkova [= С ar p in i f о 1 i a] 
crenata), lime (T i 1 i a p 1 a t у ph у 1 lu s), alder (Alnus glutinosa), 
mulberry tree (Morus andrussovi), Pitsunda pine (P i nu s pi thy us a), 
sequoia (Sequoia 1 an gs d о r fi i) and many other species. 

The list of species given above suggests that the environment of the 
flora on the shores of the Kura bay was very similar to Recent environments 
in the foothills of eastern Transcaucasia. 

Palibin (1936) and Grossgeim (1936, 1948) believed that there were no 
tropical plants in the Caucasus in Akchagyl time. The last of the evergreen 
sequoias and cherry laurels in the Shiraki area probably grew in a humid 
climate with moderately cold winters. 

The earliest occurrence ofElephas planifrons and Equus 
stenonis in the Caucasus is in Akchagyl beds, although land vertebrate 
fossils seldom appear in these beds and the material mainly consists of 
teeth of mastodon, elephant and horse. 

In Ciscaucasia, teeth of Mastodon arvernensis Croiz. et Job. 
and Elephas planifrons Falc . were collected on the Malgobek ridge, 
40 km south of Mozdok. Teeth and bones of E le ph a s planifrons (?) 
were also found in the Solenyi gulley near Grozny (Pavlova, 1931). 

In eastern Transcaucasia, the Akchagyl conglomerates between the 
Alazan and Kura rivers (Kudryavtsev, 1932) contain bones of large 
vertebrates, freshwater moUusks (Helix sp. , Planorbis, 
Cyclostoma) and species of plants known in the Recent. An antler of 
Cervus sp. is known from the Akchagyl of western Azerbaidzhan. 

The coarse breccia of the Bedeni ridge in the Tsalka District of Georgia 
contains fragmented bones of mammals among which teeth of Elephas aff. 

63 



(67) 




FIGURE 34, Fossils of the Taman complex 

1-jawofCanis tamanensis; 2- ulna of Pan thera sp. ; 3 — skull of Castor tamanensis; 
4 — upper jaw of Trogontherium cuvieri; 5 — upper jaw and teeth of Elephas meridionalis; 
6, 7 — Pmsandmetacarpusof Equus aff. siissen born ensis; 8 — jawofSus tamanensis; 9 — horn 
ofEucladocerus sp.; 10 — horn peduncle of T a m a n a 1 с es sp. ; 11 — horn axis of В iso n sp. 



64 



planifrons Falc. and Equus stenonis Cocchi have been identified. 
The breccia is underlain by a layer of doleritic lava and is covered by 
lacustrine sands and clays. The lake sediments are overlain by a layer 
of dolorite (Zaridze and Tatrishvili, 1948). Thus, in that area, the mammals 
lived and died during a period when the volcanos in the Lesser Caucasus 
were dornnant. 



68 



->- N 



d - 



f < 



h ■ 



^.AA.>.>J> -Jn,*^f^^v-»L*.-.->4-W-vn-Wtv^ 




;'<,•.•''' 



:o:.\-:e:.--i 



•'•p.;-. .о- •«.•.. e". ;.■.•.*• o*. ■•.•.; 
o'. ; • o". ' .'i^-'.'-o'".^ ''•',' ■'•'■o'.'- o." 



•.b. •■.•<>■.• 



4>- ■ о 



■b-fi.: 



c?o 



'S::^^^^ 




• ' ■ ' •\ ' •. ' • ' .'^'.' ^ .'. ' ^ ' ^^ ^ \'^ ' .'■.'^.'■'У^'. 



OS 
OB 



45 



5Л 



FIGURE 35. Stratigraphic section of 
bone-bearing sands near Georgievsk 

a- с — gravels with obsidian fragments on 
the high terrace; d-i — light- colored and 
ferruginous sands with rounded fragnnents 
of volcanic ash, bones of elephants and 
hoofed mammals. Numbers on the right 
indicate thickness in meters 



The next transgression in the Caspian 
Basin, a somewhat smaller one, is known 
as the Apsheron sea. 

The Kura bay of the Apsheron sea 
reached the longitude of Kirovabad. The 
Terek bay was temporarily connected with 
the Euxinic basin by a strait in the Manych 
area. The sea reached the latitude of 
Sarepta and [Lake] Inder in the north. 

The climate and landforms of the 
Caucasus in Apsheron time probably 
remained the same as in the Akchagyl, and 
the volcanic activity was of about the same 
intensity. Torrential mudflows, caused by 
heavy rains, carried volumes of gravel 
and boulders from the mountains 
(Kudryavtsev, 1933); these boulders can 
now be seen on the Kakhetia Plain. 

The land vegetation known from the 
Apsheron deposits in the Shiraki Steppe 
consisted of spruce (Picea orientalis) 
and a number of Recent forms: beech, 
oak, aspen, apple, willow, filbert, Turkish 
filbert, walnut (Juglans regia), 
zelkova, honeysuckle; and Hyrcanian forms: 
oak (Quercus castanei folia), alder 
(Alnus subcordata), maple (Acer 
ibericum) (Palibin, 1936). Cooling of the 
climate in Apsheron time resulted in the 
disappearance of the subtropical evergreens. 

The wooded landscape of the shores of 
the Kura bay was probably similar to those 
parts of the present-day landscapes of 
Colchis and Asterabad which are still in a 
completely natural state. 

During the regressions of the Akchagyl 
and Apsheron seas, arid conditions 
undoubtedly prevailed in the eastern part 
of the Caucasian landmass. The 
development of coastal steppes and 
semideserts created the environment for 
dispersal of steppe and desert mammals. 

Fossil mammals are more abundant 
in the Apsheron than in the Akchagyl 
sediments. The material consists mostly 



65 



of single limb bones and teeth of proboscideans and perissodactyls. These 
burials occur primarily in littoral marine deposits formed in quiet bays 
at various basin levels. Some occurrences are known from continental 
deposits: from the zone of accumulation at the piedmont "barrier" at a 
mountain base, and from the travertines in the Pyatigor'e area. 

In eastern Ciscaucasia, "traces" of a faunal complex, similar to those 
Df the Psekups and Taman, occur in the travertines and loams of Mount 
Mashuk, and in the coastal loams of the Apsheron basin. Bones of elephant 
and hoofed mammals from the travertines are brittle and some contain 
cavities formed by dissolution, and filled with calcite crystals. From 
the oldest. Upper Pliocene "travertine complex" in Mount Mashuk, the 
following forms have been described: Elephas meridionalis Nesti, 
69 E. cf. antiquus Falc, E. aff. trogontherii Wust, Bovinae, Cervinae 
(Egorov, 1932; Ivanova 1948; Gromov, 1948). 

We collected teeth of Mastodon arvernensis Croiz. et Job. , 
southern elephant (Elephas meridionalis Nesti), horse (E q u u s 
cf. s t e non i s) and bones of deer (Eucladocerus sp. ) (Figure 36) 
from alluvial gray sands below a layer of redeposited basalt fragments and 
volcanic ash (Figure 35), near the Cossack village of Vinogradnaya and 
near Georgievsk. 

Bones of Elephas meridionalis and E q u u s sp. have been found 
in the Apsheron loams on the slope of Mount Tash-Kala near Grozny 
(Pavlova, 1931). Two teeth of E q u u s stenonis Cocchi and an antler 
of a large antelope (Bub alls sp. ) found near Verkhni Achaluki on the 
Sunzha ridge are probably also of Apsheron age (Burchak-Abramovich, 
1952a). 

The following forms from the locality near the village of Vinodel'noe, 
northeast of Stavropol, are identified by Borisyak and Belyaeva (1948) in 
the PIN collections: 

Proboscidea 
Elephas meridionalis Nesti 

Perissodactyla 
Equus (stenonis?)sp. 

Artiodactyla 
Paracamelus gigas Schloss. 

A camel mandible from this locality has been described by Khaveson 
(1954b, Figure 37). 

Teeth of southern elephant from the piedmont gravels near Makhachkala 
have been recorded occasionally. 



Apsheron Peninsula 

Fossil marine and land mammals are very abundant in the limestones 
of the Apsheron stage on the Apsheron Peninsula near Binagady, Shikhovo 
and Baku. 



66 



(70) 




FIGURE 36, Fossils from the river sands in the Georgievsk quarry 

1,2 — M of Elephas meridionalis and worn surface of tooth; 3, 4 — worn surface and inner 
surface of Pmsof Equus stenonis; 5 — metacarpus of E u с la d о с eru s sp. 



67 



71 



According to Bogachev (1923, 1938a, 1940d), the fauna includes 
porpoise (Delphinus delphis L. ), an unusual predacious marine 
pinniped similar to otter and seal (Necromites nestoris Bog. ) and 
perissodactyls (Hipparion crassum Gerv. , Hipparion sp. ). The 
humerus of a seal, indistinguishable from that of Phoca caspica Gmel. 
has been found under a 22-m-thick bed of Apsheron limestone near 
Baladzhar. The presence of Black Sea porpoise (Delphinus delphis L. 
in the Caspian region can be accounted for (Bogachev, 1939a) by migration 
from the Mediterranean in Akchagyl time, i.e. , at the time of migration 
of the mollusks, Mactra, Cardium and Potamides. According to 
Bogachev, the dolphin became extinct in the Caspian at the end of the 
Apsheron because of advancing glaciation. 

The same can be said of the incompletely studied Necromites which, 
according to Bogachev (I940d), is similar to S e m a n t о r macrurus 
Orl. from the Pliocene deposits of eastern Siberia. 

Farther west, fossils occur in the clayey ridges of Kabristan and in the 
two or three parallel ridges of hills which extend from the lora and 
Kartalinia plateaus to the Gerdyman-Chai canyon in the east. The hills 
are made of clays and loams with gravelly intercalations totalling 400 m 
in thickness . 

The material collected by geologists in the Kondzhashen, Palan-Tikyan, 
Karadzha, Boz-Dag, Chuzgun-Tapa and other ridges has been identified 
by Bogachev (1938c), Burchak-Abramovich (1951a, 1952c), and the author. 



Rodentia 

Trogontherium cuvieri Fisch. 
Muridae gen . 

Proboscidea 

Aiiaiicus arveriiensis Croiz. etjob. 
Elephas meridionalis Nesii 



Perissodactyla 

E q u u s sp . 

Rhinoceros cf. eirusciis Falc. 

Rhinoceros sp . 

Artiodactyla 

С er V us sp. 

H u с 1 a d о с e ru s sp. 

Bison sp. 




FIGURE 37. Jaw of Pa ra ca m elus gigas from Vinodel'noe 



68 



74 



It is possible that the upper jaw of Stegodon sp. in the collections 
of the Natural History Museum in Baku came from some Upper Pliocene 
deposits in eastern Transcaucasia (Bogachev, 1935). 

The most abundant Upper Pliocene species from the Apsheron beds of 
Transcaucasia are, as a whole, very similar to the post-Kuyal'nitsk fauna 
of eastern Ciscaucasia. The mastodons occurring in the Apsheron beds 
were either the last survivors of the eastern Mediterranean group, or they 
were redeposited from older beds. 

Upper Pliocene index mammals have been identified from near Leninakan 
and Erzurum in the Armenian-Anatolian highlands of southern Transcaucasia. 
The fossils probably occur in the gravelly-pebbly sands deposited by streams 
in small, inland, freshwater lakes . 

As early as 1899, Abich mentioned an accumulation of fossil bones in Upper 
Tertiary calcareous marls and conglomerates underlying the layer of tuff 
and lava on the slopes of a fortified hill near Leninakan. Academician 
I.F. Brandt has identified "Mastodontidae, Cervidae and Bovidae" in that 
material. 

Bogachev (1923-1924) mentioned the finds of Elephas armeniacus 
Falc. in gray volcanic sands near Leninakan. The species is probably 
identical with E. meridionalis Nesti. 

In later collections from that area Burchak-Abramovich identified bones 
of Elephas sp. , Rhinoceros mercki Jaeger, and Equus 
stenonis Cocchi (Avakyan, 1948). 

The following list (which requires revision) is given by Bogachev (1938c) 
for species occurring in the Upper Pliocene diatomite near the village of 
Nurnus on the Zanga River: 



Carnivora 

Mustela filholi Gaudry 
P u t о r i u s sp . 

Lagomoфha 
Lepus sp. 



Rodentia 
Gerbillus sp. 

Perissodactyla 

Hipparion sp. 
Rhinoceros etruscus Falc. 



Artiodactyla 
Tragocerus sp. 

The diatomite, 7 m thick, occurs between andesitic lavas below and basalt 
above (Gambaryan, 1934), indicating that a freshwater basin existed in the 
middle Zanga region in a period between two strong volcanic eruptions. 
Intercalations of pumice and ash in the diatomite bed show that the volcanic 
activity also continued during the deposition of the bed. 

Occurrence of gerbil bones indicates that xerophilous vegetation thrived 
along the southern margins of the Armenian Highland in Upper Pliocene time. 
The stratigraphy of the fossil localities and the species composition of the 
Upper Pleistocene Transcaucasian fauna indicate that in Apsheron time the 
climate was temperate, and the volcanos were very active. These data also 
reflect to a certain degree semi-arid to arid conditions in the southern 
and eastern parts of the country. 

Pleistocene elements are traceable in the Upper Pliocene faunal complex 
of the Caucasus, particularly in the Taman Peninsula. 



69 



[12) 



TABLE 5 . Stratigraphic distribution of the main localities of Tertiary mammals in the Caucasus 



Epoch 


Division 


Stage 


Localities: Black Sea Area and Ciscaucasia 




Upper 


Chauda beds 
Gurie beds 


Taman Peninsula; Sinyaya gulley, Sennaya, 
Kuchugury 

Stavropol area: Voskresenskaya, Trans-Kuban 
Plain: Psekups, Bakinskaya 


Pliocene 


Kuyal'nitsk 


Taman Peninsula: Kapustina gulley 




Middle 


Cimmerian 


Zmeika ridge: Darg-Kokh 




Lower 


Pontian 


Stavropol Plateau: Kosyakin quarry 


Miocene 


Upper 


Meotian 
Sarmatian 


North Azov coast: Novocherkassk, Tuzlovka 

Stavropol area: Burlatskoe. Blagodarnoe, 
Petrovskoe 




Middle 


Konka 

Karagan 

Chokrak 

Kuchugury 

Helvetian 


Stavropol area: Belomechetskaya 




Lower 






Oligocene 









70 



;7з: 



Stage 


Localities: Caspian Area and 
Transcaucasia 


Predominant Groups of Species 


Apsheron 


Terek Range: Malgobek, Grozny, 

Achaluki 
Apsheron Peninsula: Binagady, Khurdalan, 

Shikhovo 
Armenian Highland: Leninakan, Nurnus 


Taman wolf, southern elephant, 
T r go n t h er i u m , Equus 
stenonis, Cervus pliotaran- 
doides , Taman bison 


Akchagyl 


lora Plateau: El'dar, Dzheiran-Cheli 
Kirovabad Plateau: Karasakhkal, Naftalan 
lora Plateau: Palan-Tikyan 


Mastadon arvernensis 

Aceratiierium 

H i p p a r i П 

Cervus pliotarandoides 


Productive beds 


Apsheron Peninsula: Lok-Batan 


Fox, 
deer, 
gazelle 


Pontian 




D i n с у n , 

Mastodon arvernensis, 
hamster, D i с e r or h i n u s , 
tapir, Hipparion, roe deer 


Meotian 
Sarmatian 


Kartalinia Plateau: Dzhaparidze 
Gori valley: Arkneti 
lora plateau: El'dar, Udabno 
Kartalinia Plateau: Kyasaman, 
Kotsakhuris-Kedy 


Hyena, long-snouted mastodon, 
Aceratherium. Hipparion, 
giraffe, gazelle 


Konka 
Karagan 
Chokrak 
Helvetian 




A m p h i с у n , swamp mastodon, 

P a г a 11 с It i t h e r i u m , 

A n с h i t h e r i u m , giant swine, 

antelope 










Dzhavakhetia: Akhaltsykh 


1? e n a r a t h e r i u m 
Chalicotherium 
A n t h r a с t h e г i u m 
Tragulids 



71 



The Neogene faunal complexes of the Caucasus are not endemic. The 
Oligocene "fauna" of the Lesser Caucasus is related to the central Asian 
and European "faunas" of the same period. 

The Middle Miocene "fauna" of Ciscaucasia, known from Belomechetskaya, 
is also related to the Miocene "faunas" of Eurasia and North America. This 
"fauna" migrated over temporary land bridges from the north and south, 
although the trend from the south was more pronounced. 

In the Upper Miocene, Sarmatian, the Hip par ion complex migrated 
to the Caucasus from the south over land which joined the island of 
Caucasus with Asia Minor. 

The local fauna of the mountains has probably evolved since the Miocene. 
Some of the surviving, endemic elements of the highland fauna are 
undoubtedly descendants of Upper Miocene forms. 
75 Evolution of the Pliocene faunal complexes in the piedmont and the 

plateaus can be traced mainly to migrations from the north and south. 
The hot, dry climates of the Pliocene produced a more pronounced southern 
influence on the evolving fauna. 

It is only at the very end of the Pliocene that the northern faunal elements 
became important. So-called Pleistocene genera and species appear during 
the general cooling at the end of the Pliocene. Evolution of endemic species 
in the Pliocene occurred mainly in the highland regions and left relicts on 
the peneplain. 

The paleontology of the Caucasian land mammals does not support the 
hypothesis of a "Pontian landmass" in the Black Sea Basin connecting 
western Caucasia, the Crimea and Asia Minor (Andrusov, 1918; Vul'f, 
1944; Puzanov, 1949). This hypothesis has also been discarded by geologists 
(Muratov, 1951). However, it is more than likely that dry land existed in 
the Upper Pliocene where the Sea of Azov now lies (Vereshchagin, 1957). 

A stratigraphic summary of Caucasian localities of Tertiary land 
mammals is given in Table 5. 

Many forms of the Upper Pliocene Taman complex survived in the 
Caucasus and continued to evolve through the Lower and Middle Pleistocene. 



72 



76 Chapter II 

DEVELOPMENT OF CAUCASIAN LANDSCAPES AND 
MAMMALIAN FAUNA IN THE QUATERNARY 

THE PLIOCENE — PLEISTOCENE BOUNDARY 
IN THE CAUCASUS 

The specific features of evolving landforms in this area between two 
ancient continents make the identification of the Pliocene-Pleistocene 
boundary difficult. 

The variety of zoogeographic zones and the complexity of their boundaries 
developed in the Upper Pliocene as well as in the Recent. 

Therefore, any study of the geomorphology based upon single occurrences 
of plants and animals in continental deposits often reflects only local 
conditions. 

This problem, as it concerns the European continent, is still under 
discussion by geologists and biologists (Zhirmunskii, 1936; Nikolaev, 1947, 
1950). We agree with Nikolaev that the Pliocene -Pleistocene boundary must 
be identified on the basis of all available data of the various disciplines, 
rather than on isolated criteria. 

A variety of geomorphological changes caused by the interplay of 
tectonics and climate have been proposed as the markers of the Pliocene- 
Pleistocene boundary. Such markers are sea level fluctuations, orogenic 
and erosional cycles, glaciations, evolutionary changes of plants and 
animals, and morphogenetic stages of fossil lineages. 

The lower boundary of the Quaternary has been variously placed — in the 
Upper Sarmatian by geologist Kovalevskii (1936) and botanist Grossgeim 
(1936); at the end of the Akchagyl by geologists Reingard (1931, 1936a), 
Gerasimov and Markov (1939) and paleontologist and geologist Gromov 
(1948); and in the Upper Apsheron by geologists Pavlov (1936), Mirchink 
(1936b) and Vardanyants (1948). 

The Russian zoogeographers Menzbir (1934) and Serebrovskii (1935) 
did not resolve the question when they applied Western European 
stratigraphic concepts to the Caucasus. 

The hypothesis of Kovalevskii, Grossgeim and Shvetsov (1928) that the 
Pliocene continental glaciations in the Caucasus correlate with the Giinz 
and Mindel glaciations is no longer accepted. 

Gromov (1948) inferred from the geomorphological and paleontological 
data that, some time at the Tertiary-Pleistocene boundary, the climate' 
became more severe which, however, did not imply an extensive glaciation. 
He considered the Khapry and the Psekups faunal complexes (see Chapter I) 
characteristic of the transition from the Pliocene to the Pleistocene in 



73 



77 



Ciscaucasia and the southern part of the Russian Plain, and placed the 
Taman complex at the base of the Quaternary. * 

At that time the following forms were considered indicative of early 
preglacial time on the Caucasian Isthmus: Elephas trogontherii 
(E. wusti M. Pawl.), Bison schoetensacki Freud., Elasmo- 
therium caucasicum Boris., Cricetus cricetus L. , Spalax 
m i с r о p ht h a Im u s Giild. The faunal break was expressed in the 
disappearance of mastodon, elephant (Elephas planifrons and 
E. meridionalis), Hipparion and Etruscan rhinoceros, coupled with 
the appearance of bison. Thus Gromov's biostratigraphic criteria of the upper 
margin of the Tertiary are similar to those of West European authors: the 
Khapry and Psekups complexes coinciding with the Giinz and Gunz-Mindel 
glaciations, and the Taman complex with the Mindel glaciation of the West 
European glaciological scheme. 

The first erosional cycle in the Quaternary occurred in the Giinz. The 
cycle is represented by gravels of the Kuban River terrace which are 
175m thick. 

Similar paleontological considerations have been put forward by 
Pidoplichko (l940c, 1952, 1954) for the northern Black Sea coast and the 
Ukraine. 

New paleontological data have confirmed that pronounced qualitative 
changes in the flora and fauna of the Caucasian Isthmus occurred only at 
the end of Apsheron time. These changes are indications of a mildly warm 
or temperate climate. 

Silicified wood of broadleaf species — oak (Q u e r с u s sp. ), beech 
(Fagus sp. ) and elm (U 1 m u s sp. **)— has been recorded from Upper 
Pliocene freshwater beds near Kuchugur on the Taman Peninsula, and in the 
lower part of the alluvial sands in the Girei quarry on the Kuban. It is 
quite possible that subtropical species did not exist in the Caucasian flora 
of this period. An abundance of terrestrial mollusks (Chondrula, 
He 1 ix) and freshwater mollusks (Corbicula, Anodonta, Bithynia, 
Micromelania and other species) in the post-Kuyal'nitsk continental 
beds of the Taman Peninsula and in the Upper Apsheron beds of eastern 
Transcaucasia, is evidence of a temperate climate. 

It was in Upper Apsheron time that mastodon and Hipparion became 
extinct in the foothills of the northern Caucasus. The forms which then 
appeared were horse (Equus cf. stenonis, E. aff. siissenbornen- 
sis), E 1 a s m о t he r iu m caucasicum, camel (P a r a с am e lu s) 
and bison (Bison sp. ). 
78 The morphogenesis of the Caucasian elephant (E le ph as meridiona- 
lis — E. t r о go nt he r i i) was in a transition stage during the Upper 
Apsheron (Vereshchagin, 1957a). 

The picture is similar in the continental deposits of the Transcaucasus, 
particularly so in the Armenian Highland. However, the chronological 
correlation with the Ciscaucasian localities is uncertain, despite the 
similarities in the composition of the faunal complexes and in the 
morphogenetic stages. 



When Gromov wrote this, the Taman complex was poorly known; in point of fact, it is very similar to 
the Psekups and Khapry complexes (Vereshchagin, 1957a). 
Identified by A.I. Zubkov. 



74 



Thus at the end of Apsheron time, new cycles of erosion and mountain 
formation commenced on the Caucasian Isthmus; the flora developed a 
temperate or, in some places, a xerophilous aspect, and the vast areas 
frona which the seas receded came to be inhabited by ungulates of a type 
adapted to the steppe, or even to the desert. 

Taking into account that this continental phase undoubtedly covered a 
very long period, we agree with Pavlov (1936) in placing the beginning 
of the Anthropogene in the Caucasus at the end of the Apsheron sea 
regression and at the beginning of the Baku sea transgression. 



THE PROBLEM OF THE CAUCASIAN CLIMATES AND 
LANDSCAPES IN THE PLEISTOCENE AND THEIR BEARING 
ON THE EVOLUTION OF FAUNAE COMPLEXES 

The chief considerations in studying the development of the climates and 
landforms of the Caucasus in post -Apsheron time are the following: 

1. Tectonic cycles and the dynamics of the land relief created by river 
erosion. 

2. Fluctuations of the sea level and the land connections between the 
Ciscaucasus and the Russian platform. 

3. The extent and number of glaciations as related to the tectonics and 
climatic fluctuations . 

4. Changes in flora and formation of vegetation zones. 

These factors have been studied in the Caucasus on an empirical basis, 
disregarding the paleontological material. A summary of these studies 
would be germane to the overall problem of the history of the fauna. 

Dynamics of land relief. In the Pleistocene the dynamics of the 
relief of the Caucasus was highly complicated. The peneplanation of the 
ancient mountains of the Greater and Lesser Caucasus at the end of the 
orogenic period of the Top Pliocene (Apsheron age) is a fact accepted 
by geomorphologists. 

Remains of an ancient peneplain are in fact observable in the Main 
Range, e.g., in Abkhazia, Svanetia and Kabarda, and particularly in 
Dagestan. At the end of the Pliocene, the poorly drained, boggy alpine 
plateaus of the present Dagestan highlands were probably no higher than 
500-800 m above sea level. The peneplain is considered to be of either 
Apsheron or Baku age; Mirchink ( 1936a) regards it as Upper Apsheron- 
Lower Baku. Some ridges in the region of Vodorazdel'nyi and the Bokovye 
ridges presumably did not rise over 1,000-1,500 m above the level of the 
Apsheron sea. Plant and animal fossils from these plateaus would have been 
79 of great interest in the study of the faunal complexes. Unfortunately, their 
preservation was probably precluded by the action of rapid drift and erosion. 

According to Nikolaev (1941, 1949) the relief of the inner plateaus of 
the Greater Caucasus is much older than Apsheron time. The relief of the 
Armenian Highland (Paffengol'ts, 1948, 1950) can be traced as far as the 
time in the Lower Quaternary when Lake Sevan was formed by the lava 
flows from the Bogu-Dag volcano which dammed the deep erosional valley 
of the paleo-Zanga River. 

Nikolaev (1941, 1949) places the maximum uplift of the Greater 
Caucasus, following a series of oscillating, tectonic movements, at 3-4 km. 



75 



which was reached in "post-Upper Pliocene" time. On the basis of their 
studies in Dagestan and Ossetia, Reingard (1939) and Vardanyants (1933b, 
1948) hypothesized that the central part of the Main Range is a horst, 
which was uplifted by 1,000-1,500 m. The figure of 3-4 km is probably 
the sum of the uplift of the Central Range and the subsidence in the Black 
Sea and Caspian Sea depressions. 

The most recent tectonic movements have been detected in the central 
Caucasus, in the Stavropol area and on the Taman Peninsula. The laccoliths 
in the Pyatigor'e area, which stand out in the surrounding relief, "underwent 
a relatively greater uplift than the surrounding areas. " Ronov (1949), 
working by a "volume method, " stated that the uplift of the Greater Caucasus 
in the Pleistocene was nearly 2,000 m. 

Thus, by a general consensus, the relief of the Caucasian mountains 
resulted from arching, or, in some places, from horst uplift, and from a 
concomitant subsidence in the piedmont plains and in the Black Sea and 
Caspian Sea depressions. 

Because of these tectonic movements, the younger terraces in the 
foothills were formed close by older, denudational terraces, which, in the 
plains, disappeared completely under the younger ones. This overlapping 
of terraces is very important in a stratigraphic correlation of terrace 
deposits and in fossil dating. 

The uplift of the Caucasian mountains inevitably resulted in glaciations 
(which are discussed in more detail below) and in the formation of a rocky 
relief with numerous cuestas and talus in the faulted areas. The new 
topography allowed an expansion of plant and animal species adapted to a 
rocky terrain. The higher relief also promoted a greater ecological and 
morphological differentiation which was reflected in the formation of new 
habitats and in the division of existing ones. 

The slopes of the ridges which rose during the Pleistocene were actively 
eroded by rivers. Mountain streams deepened their valleys, stage after 
stage, forming terraces which correspond to the mountain-building cycles. 
Rivers cut through the young rising ridges forming narrow canyons in the 
process. 

Such post-Pliocene canyons occur on the northern slope of the Main 
Range (along the Baksan, Chegem, Cherek and Urukh rivers); in eastern 
Transcaucasia (along the Geok-Chai, Tur'yan-Chai, Alazan, Kura and 
Mtskheti rivers); on the southeastern spurs of the Karabakh Mountains 
(along the Araks River), and in Talysh (along the Vazaru-Chai river). 

The differential uplift of the mountain ridges combined with river erosion 
produced sheltered valleys, screened by ridges, which captured atmospheric 
precipitation and resulted in the development of temperate, arid areas in 
the highlands. These were a refuge for plants and animals adapted to 
drought. During the Pleistocene such areas existed in Dagestan, in the 
longitudinal mountain valleys of Ossetia and Kabarda, in the Gori depression 
on the Kura River, in the upper part of the Pambak River valley on the 
Armenian Highland, and in the Zuvanda depression in Talysh. These areas 
are still inhabited by xerophilous animals: suslik, hamster and steppe vole. 

Karstic niches and caves which were created in limestone terrain by 
river erosion became habitations for some predator species and, later, 
for man. 

A thick mantle of gravels and loams formed in the piedmont during the 
Pleistocene from the transport of clastic material by rivers from the 



76 



mountains. Such deposits cover the Trans -Kuban and Terek-Sunzha valleys, 
the Kusary and Kirovabad plains in Transcaucasia, and the terraces on the 
Black and Caspian seacoasts. 

These piedmont plains were inhabited by mammals associated with the 
forest-steppe and with moderately humid climatic conditions. The piedmont 
plains were also the paths of migration from north to south, and vice versa, 
of those species which lived on the lowland plains. 

Vulcanism has had a pronounced effect on the relief and hence on the 
areas of species distribution in the highlands and on the piedmont plains 
(Apsheron and Taman peninsulas). In the Armenian Highland the lavas 
covered great areas of the plateau, sometimes nearly 50% of a whole area 
(Maruashvili, 1946). Volcanism affected the distribution of animal species 
both directly and indirectly. For example, the absence of forests in western 
Armenia is, according to some geobotanists, the result of the volcanic 
activity and lava flows of the Pleistocene (P. Yaroshenko, 1941). 

The orogenic cycles were probably closely related to fluctuations in 
sea level during the Quaternary. 

Fluctuations of sea level. The transgression of the Baku sea, 
which followed the Apsheron regression, was characterized by an abundance 
of Didacna. The next stage, the Khazar sea, was characterized by the 
archaic Didacna surachanica Andr. The equivalent in the Black Sea 
Basin was the Neo-Euxine sea. The Khazar sea was followed by the 
brackish Khvalynsk sea which extended far to the north leaving brackish- 
water mollusk fauna near Ural'sk and Kazan. The Khvalynsk sediments 
81 correlate with the beds of the Ancient Black Sea. The last major 

transgression in the Caspian depression (Khvalynsk sea) is usually thought 
to correspond to the last stage of glaciation of the Caucasian Range. The 
Khvalynsk sea receded into the postglacial Caspian, in which later trans- 
gressions were only minor and did not extend much beyond the present 
coastline (Kovalevskii, 1933; Figure 38). 

The major Pleistocene transgressions considerably reduced the coastal 
terraces on the east and west coasts of the Isthmus. 

Sea erosion of the piedmont barrier, as in eastern Transcaucasia, or 
of the coastal plains restricted the areas of distribution and migration of 
animals on the plains. The alluvial plains were also affected by marine 
erosion: the Yeisk-Kuban, Kuma-Terek, Rion and Kura-Araks plains, 
which, at various times, were inhabited by faunas of diverse origins. 

The Manych strait formed during the transgression must have restricted 
the migration of animals from the Russian plains at least to the extent that 
hibernating animals could not have migrated over ice -covered country. 

During the regressions the low -lying areas remained covered by broad 
Salinas and marshes with residual lagoons and lakes. Desalinization of 
these areas progressed rapidly in western Ciscaucasia where steppes 
32 developed. In eastern Ciscaucasia, desalinization proceeded at a slower 
rate. There semideserts developed, inhabited by jerboas, gerbils, saigas 
and goitered gazelles. Residual lakes and lagoons became plavni overgrown 
with nymphaea and bulrushes. Colchis developed a marsh-type vegetation 
consisting of alder, nymphaea, sedge and sphagnum. The longevity of such 
marshy tracts depended on the hydrological regime of the feeding rivers 
and the rate of alluvial fill. The fauna of these marshes included such 
hydrophilous animals as otter, water rat, European beaver and boar. 



77 



(81) 




Asterabad 



FIGURE 38. The Caucasus and the Caspian in the Upper Pleistocene 

1 — old freshwater sediments in the Volga Basin; 2 — Aral-Caspian marine 
transgression; 3 — region of Quaternary glaciation (from Kovalevskii, 1933) 



Extent and number of glaciations. Climatic fluctuations and 
glaciations in the Caucasus could have been the most important factors in 
the development of the fauna. 

Data on Caucasian glaciations is based on the geomorphology of the area 
and on the relationships between the glacial and orogenic cycles and the 
sea level stands. Correlation of these data has been complicated by the 
attempts of Russian geologists to identify Caucasian glaciations with the 
European glacial stratigraphic scheme developed for the Alps. There has 
been almost no original stratigraphic work done to date on the glaciations 
in the Caucasus that is free of this "Western glaciological hypnosis. " 
Geomorphological studies show those peculiarities which distinguish the 
development of the Caucasus during the Pleistocene from the development 
of the Alps. 

Academician Pavlov (1925) believed that the end of Apsheron time in the 
Caucasus coincided with the Pliocene (Gilnz) glaciation, and that it was marked 
by the uplift of ranges causing a formation of mountain glaciers and intensive 
deposition of the gravels which are common in the upper part of the 
Apsheron suite. According to Pavlov, the Mindel glaciation correlates 
with the upper Baku and the Riss glaciation with the Upper Khazar, while the 



78 



Khvalynsk corresponds to the time when the Wiirm glaciers waned. 
Essentially this correlation is still valid. 

Reingard (1937, 1947b) recognized four or five major stages of glaciation 
in the Caucasus. The first (Giinz), which glaciated as far as the piedmont, 
occurred in the transition from Akchagyl to Apsheron time, and is 
represented only by scattered gravel deposits. The Mindel glaciers extended 
80 km north on the piedmont plain of the central Caucasus during the 
Apsheron-Baku transition period. This glaciation was followed by the long 
second interglacial, during which thick sequences of diluvial loams 
accumulated from transport of clastic material from the mountains. The 
third glaciation, the Riss, was the most extensive and its erosional effects 
in the valleys were the most pronounced. During the interglacial which 
followed, the glaciers remained only on the mountain peaks. During the 
fourth and the last glaciation, the glaciers covered only the nnountain 
valleys, and subsequently receded in four stages, with one advance which 
occurred in the Middle Ages of our era. 
!3 According to Reingard, the rise in the level of the Caspian Sea was 

related to the decrease in evaporation at the time of glaciation. Initially, 
the sea level rose slowly. Later, during the melting of the glaciers, the 
rise was more rapid, attaining its maximum in the second half of the glacial 
phase. At that time, the Black Sea level had not yet reached its maximum 
(Figure 39). 

According to Gerasimov and Markov (1939), the entire glacial history 
of the Caucasus can be subdivided into two periods. During the first period 
(Akchagyl-Apsheron), one or two glacial phases occurred in central and 
eastern Caucasia as the result of the uplift of the mountains. In the second 
period (the second half of the Quaternary), two glacial phases developed in 
the Greater Caucasus, "the mountain framework formed by the Lower 
Quaternary tectonic movements. " The authors doubted any synchrony 
between these and the alpine glaciations, although they were certain of 
"basic changes in climate which occurred several times, along with a 
number of glaciation (3-4) in the Caucasus. " 

A record number of ten glaciations beginning with the Upper Miocene 
has been given by Kovalevskii (1936). 
84 Three or four glacial phases in the Caucasus have been suggested by 
Vardanyants (1948). However, he tended to accept Gromov's (1935b) 
subdivision into two glacial epochs corresponding to the Riss and Wiirm of 
Europe: the Riss corresponding to the Early Khazar, and the Wiirm to the Early 
Khvalynsk, with corresponding stages of retreat following each (Figure 40). 

In the Black Sea Basin the periodically established connections with the 
ocean were controlled by the uplift and subsidence in the Bosphorus- 
Dardanelles region and modified the processes of glaciation and 
transgression. A stratigraphic summary of Vardanyants' is given in 
Table 6 . * 

A number of marine basins periodically connected the Aral, Caspian, 
Black and Mediterranean seas, and their waters interchanged with the 
fresh waters of the Caspian through the Manych strait. Increase in salinity 
of the basins was due to the penetration of the Mediterranean water eastward 
in post-Tertiary (Karangat and Ancient Black Sea) time. 

* Other geologists, for example Popov (1955), correlate the Khvalynsk time with the Neo-Euxine beds, 
the Khazar time with the Karangat beds, etc. 



79 



It seems strange that the views of geologists on the extensive glaciations 
in the Caucasus were supported, rather than opposed, by biologists. 



:8з; 




FIGURE 39, Relationships between glaciations, otogenic phases in the Caucasus and fluctuations 
of the Caspian Sea level (according to Reingard, cited in Bogachev, 1940) 

Formation of vegetation zones. According to the botanists 
Kuznetsov (1909) and Medvedev (1915) and the zoologists Satunin (1910) 
and Dinnik (1911), there was only one glacial phase during which the 
Caucasian fauna and flora were saturated with northern forms, and the 
mountain animals and plants had to seek refuge on the plains. 

From an analysis of Recent Caucasian flora, Grossgeim (1948) has 
inferred a number of hot interglacials, in order to account for the presence 
of xerophilous southern plants at a number of places in the Caucasus. 
Geobotanists Maleev (1946) and Kolakovskii (1947b), who studied the origin 
of the eastern Mediterranean flora (in particular, that of Colchis), also 



80 



(84) 




FIGURE 40. Glaciation in the Caucasus during the pre-Khazar erogenic phase 
1 _ inferred outline of the sea; 2 — areas of glaciation (from Vardanyants, 1948) 



accepted the notion of several glacial epochs and warm interglacials. They 
connected the occurrence of mesophilous broadleaf forests to the pluvial 
glacial periods, and the reduction of mesophilous flora, with concomitant 
development of semixerophilous and thermophilous vegetation, to the arid 
interpluvials. These conclusions are based either on studies of 
discontinuities in plant distribution, or on the geological literature 
(Vereshchagin, 1949c). An interplay of many factors, other than climatic, 
is responsible for discontinuities in the distribution of species. Even the 
penetration of northern plants (e.g., sphagnum and sundew) far into the 
south —to the Greater Caucasus, the marshlands of Colchis and the Lesser 
Caucasus highlands — cannot be correctly evaluated unless all the factors 
which control species distribution are known. The use of phytocenotic 
transformations to detect climatic changes is open to question even when 
applied to a period of time as short as the historical epoch (e.g., Sinskaya, 
1933; Fedorov, 1952). The Pleistocene climatic changes cannot be evaluated 
from the fossil plant record alone because the number of known macroscopic 
plants is small. However, the available paleobotanical and geobotanical 
evidence indicates that the flora of the Caucasus in the Middle Pleistocene 
resembled the Recent flora and the plant habitats had the essential aspects 
of the Recent. The arboreal vegetation at the end of the glacial epoch was 
also similar in type and in distribution areas to that of Recent time 
(Grossgeim, 1936, 1948). 

The valley glaciers of the past, as, for instance, those on the slopes 
of western Caucasia, were much larger than those of the Recent. This is 



cor\TABLE 6. Stratigraphic summary of post 
^ ^ Sea (from Vardanyants, 1948, p. 23 ) and 



-Tertiary history of the Bla 
connections via the Bospho 



ck Sea, the С 

rus and the M 



aucasus and the Caspian 
anych region 



Bosphorus 


Black Sea 


Caucasus 


M anych 


Caspian Sea 




Recent sediments 


Bijhl stages 


Open ( ?) 


Post-Khvalynsk layer 


Closed 


Subsidence, regression, 
decreasing salinity 


Uplift, glaciation (Biihl 
or Neo- Wijrm ) 


Closed 


Regression 




Ancient Black Sea beds 


Wijrm stages 


Open 


Khvalynsk layer 


Open 


Subsidence, influx of 
seawater 


Uplift and glaciation 
(WiJrm) 


Closed 


Regression 




Neo-Euxine beds 


Riss-Wiirm 


(?) 


Khazar layer 


Closed 


Subsidence, regression, 
decreasing salinity 


Uplift and glaciation 
(Riss) 


Closed 


Regression 




Karangat beds 


Mindel- 
Riss 


Open 


Upper Baku (transitional 
beds) 


Open 


Subsidence, influx of 
seawater 


Uplift 


Closed 


Regression 




Uzunlarskoe and Ancient 
Euxine beds 




Closed 


Middle Baku 
Lower Baku 




Chauda beds 


Uplift and glaciation 
(Mindel) 


Closed 


Regression 




Krasnodar horizon? 


Giinz-Mlndel 


Closed 


Upper Apsheron 


Closed 


Regression and decreasing 
salinity (? ) 


Uplift and glaciation 
(Gunz) 


Closed 


Regression 




Guri beds 




Open 


Lower and Middle Apsheron 



indicated by the occurrence of glacial troughs and terminal nnorainic ridges 
in the valleys of the Laba, Zelenchuk, Kuban, Baksan and other rivers in 
the area (Figure 41). 

The glacial troughs occur in the upper parts of the valleys, far short 
of the Skalistyi ridge. It is clear that there were no basic differences in 
the environments of the faunal and floral complexes between the time when 
the valleys were covered by glaciers, andthe Recent. The vegetation of 
the highlands descended, although it was not completely displaced, to the 
j7 piedmont plains. * During the period of majiimum glacial advance, there 
was adequate living space for mountain animals on the meridional and 
lateral mountain ridges near the glaciers. Even at present, mountain forests 



• Even Grossgeim (1936, p. 203), allied as he was with Kovalevskii, was compelled to recognize this fact. 



82 



(coniferous, broadleaf and mixed) and mountain animals exist close to the 
glaciers, as on the Cherek, Urukh and Rion rivers (Figure 42). 

The gravel beds and large boulders in the piedmont plains were more 
likely the result of mudflows, caused by floods and volcanic activity, than 
relicts of hypothetical glaciations. 

The evolution of mammalian faunal complexes must be related to the 
snow factor and to the latitudinal shift of the climatic zone on the Russian 
Plain caused by the northern European glaciation. However, the depth and 
duration of the Pleistocene snow cover have not been studied, and can only 
be inferred from examples of distribution and migration of some nnammals 
in the Recent. 




FIGURE 41. Glacial trough and terminal moraine, overgrown with pine, in the Tseya ravine in 
North Ossetia 

Photograph by author, 1947 

The occurrence of boreal plant species in the Caucasus can only be 
accounted for by the proximity of the Riss glacier tongues to the Caucasian 
Range (Medvedev, 1915; Grossgeim, 1936, 1948). In other words, it is 
assumed that tundra and taiga existed in the Manych area and in the Salsk 
steppes. Results of recent palynological and paleontological studies indicate 
that steppes always existed between the lower Don and the lower Volga 
rivers. 

We will evaluate the paleontological and zoogeographical data in the 
Caucasian stratigraphy after we review the Pleistocene localities. It should 
be noted that the hypothesis of extensive continental glaciation in the 
Caucasus implies that the evolution of all the highly specialized forms 



83 



(mountain goat, snow mouse and other forms) took place within a very short 
time — the postglacial period. 

The glacial stratigraphic scheme, particularly the hypothesis of 
continental glaciations in the Caucasus, should be reviewed with reference 
to biological data. 



!) 




FIGURE 42. Glacier and forest in the upper reaches of the Rion 

Photograph by author, 1957 



89 



Absolute ge о с hr о no lo gy of the Quaternary. Application 
of the absolute time scale of the Quaternary to the Caucasus did not produce 
new results. Most Russian authors accepted the estimates of Soergel (1925) 
and Milankovich (1939) of 600,000-650,000 years as the duration of the 
glacial epoch. In estimating the time of origin of the Aral-Caspian sea 
at 23,000 B.C., of the Khazar sea at 13,500 B.C. and of the Khvalynsk sea 
at 3,400 B.C., Kovalevskii (1933) condenses the entire history of man and 
the evolution of the Caucasian fauna into a very short period of time. Later 
investigators have found that the Quaternary orogenic processes and 
glaciations in the Caucasus lasted much longer — 500,000-600,000 years 
according to Vardanyants (1948) or even 1,000,000 years according to 
Nikolaev (1949) and to Ronov (1949). These estimates, which are of great 
importance in the calculation of the rates of evolution, speciation and 
development of the fauna, will undoubtedly be amended in the course of 
future studies of the physicochemical composition of rocks, bones and 
sediments. The only Caucasian stratigraphic stage with a satisfactory 
paleontological record is the Upper Pleistocene — presumably the time 
when glaciation reached its maximum extent. 



84 



MODE OF OCCURRENCE OF FOSSIL MAMMALS AT 
THE PLEISTOCENE LOCALITIES 

Pleistocene localities are mc^re amenable than those of the Tertiary to 
studies of the conditions of animal death and deposition. 

In the Caucasus Quaternary mammal fossils occur in marine terraces 
along the coasts of the Black, Azov and Caspian seas, in river and lake 
terraces, in the surface loams and in caves. 

Only a few occurrences of mammals in marine and lagoonal sediments 
are known in post -Apsheron strata. This is probably due to the latest 
tectonic processes, since it is unlikely that the variability of the animal 
world in the Pleistocene was any less than in the Tertiary. 

The bodies of animals which had died from various causes (see 
Introduction) accumulated at river mouths and estuaries. Smaller 
accumulations formed in the alluvial sediments of the middle and lower 
parts of rivers. All such sites in the lower parts of the valleys of the 
Kuban, Terek, Rion and Kura rivers are at present in the zone of tectonic 
subsidence, in which sediments continue to accumulate. Consequently, 
most of the sites remain concealed by the sedimentary cover and can be 
discovered only in the processes of drilling and quarrying. 

Exceptions to this usual mode of occurrence are found in those sites 
which were at the head of coastal bays, provided the bays were in the area 
of the later uplift of the foothills. Such localities are found in the middle 
parts of the Kuban (Kropotkin, Nevinnomyssk), on the Terek River (near 
Mozdok and Kotlyarevskaya), on the Rion (at the longitude of Kutaisi and 
Zestafoni), and on the Kura (in the Mtskheti-Tbilisi area). 

It is possible that several fossiliferous layers occur at localities in a 
"zone of undulation" (i.e., longitudinal depressions and swells) because 
favorable hydrological conditions prevailed over several epochs. 

Most of the Pleistocene mammals are found as intact or fragmented 
skeletons in continental deposits of various origins (alluvial, lacustrine, 
diluvial, eolian) in caves, travertines and asphalt pools, on the plains and 
in the foothills of the Caucasus Isthmus. 

Although life was precarious in the mountains and avalanches, falling 
90 rocks, slippery, ice-covered surfaces and hunger brought frequent death 
to the animal communities, the chances were poor for the preservation of 
their remains. The fine-grained alluvial sediments of mountain rivers 
did not offer much likelihood of preservation either . Such sediments are 
usually thin and tend to shift with time (Shantser, 1951). Fossil vertebrates 
do not occur, as a rule, in the terminal and lateral moraines of mountain 
valleys, because of the attrition of bones in the process of moraine 
deposition. 

The diluvial sediments at the foot of the mountains, far from the valleys 
of large rivers, do not contain fossil accumulations, since skeletons were 
usually destroyed during the slow deposition of the sediment. 

Skeletons of animals carried by torrential streams onto the plain areas 
stood a better chance of preservation in the thick mud and conglomerates 
formed by the streams. Mudflows continued to cause destruction and death 
among animals throughout the Quaternary; they have been vividly described 
by Ivanovskii (1911) for the Araks River valley, and by Muratov and 
Gvozdetskii (1948) for the Khasaut valley in the northern Caucasus. 
Mudflows are also known to have occurred in the rivers of eastern 



85 



Transcaucasia, for example, in the Demir — Aparan-Chai ravine. 

Mudflows are produced by such factors as heavy rains and volcanic 
eruptions which melted the firn and ice. The flows, which were more 
frequent and powerful in the Pleistocene, killed masses of animals. 

Heavy snow and surface ice were probably common causes of violent 
death among the Pleistocene animals of the plains, just as they are among 
the Recent animals. On the plains beyond the river valleys bones were 
rarely preserved and then only in the gulleys and diluvial loams. 

Disposal of food remains by primitive man at campsites and in caves 
added a new factor to the accumulation of Pleistocene fossils. The prey 
carried by predatory animals and owls also contributed to the bone 
accumulations of many species at the same cave sites. 

This type of death assemblage occurs most frequently in the mountains 
and is the main source of information on the history of the mountain fauna. 

The sections which follow are devoted to the discussion of the niajor 
localities by regions. 



PLEISTOCENE BONE-BEARING LOCALITIES 
Western Ciscaucasia 

Following the deposition of the bone -bearing Psekups gravels and 
conglomerates on the Taman Peninsula, no rich, diversified mammalian ac - 
cumulations are known for quite a long time . The Lower and Middle Pleistocene 
fossil record consists mostly of teeth ofElephas meridionalis and 
E. trogontherii. 
91 On the Kuban Plain, i.e. , on the right bank of the latitudinal section of 

the Kuban, fossil mammals occur at great depth in the sediments of ancient 
rivers which ran off the Caucasian mountains. 

The remains were deposited on the edges of floodplains of the plavni or 
liman type fed by mountain streams, probably beyond the zone of deposition 
of coarse elastics. 

The clays and sands of the Kuban Plain (e.g., those in the Krasnodar 
area) contain bones of large Middle Quaternary mammals of Khazar time, 
along with wood and freshwater mollusks (Unio, Paludina) 
(Pravoslavlev, 1932). The overlying beds are loess loams with 
occasional intercalations of volcanic ash. In addition to mammals and 
plants, these beds contain terrestrial niollusks (Buliminus tridens 
Mull. ), which are indicative of a steppe environment. 

Bones of "Bos, Equus, Cervus, Elephas" have been recorded 
by Pravoslavlev and Ansheles (1930) from the 50-m-thick loams near the 
Cossack village of Temizhbekskaya. 

A number of fossils from other localities were identified by the author 
in 1952 in the Krasnodar Museum. These include the teeth and skull 
fragments of an early mammoth, Elephas trogontherii — 
E. primigenius, from the village of Dinskaya, and remains of a long- 
horned bison. Bison priscus longicornis, from the village 
of Krylovskaya on the middle Chelbas. The skeleton was buried in the 
silty sand of a stagnant-water basin. The bones were only slightly 
permineralized and resembled alabaster in their appearance. Bones of 



86 



Bison priscus longicornis and mammoth were preserved under 
similar conditions near the village of Skobelevskaya. The bones were 
found in a silty marsh bed 16.5 m beneath the loess. Elongated grooves, the 
tracks of boring invertebrates (Trichoptera larvae ? '. ), occur on the 
surfaces of the bones . The occurrences of B. priscus longicornis 
indicate the Middle Pleistocene, Khazar age of these localities. 

Five nearly complete skeletons of small cave bear (Grigor, 1926), 
described by Borisyak (1930, 1931) as Spelaearctus rossicus, were 
found near Krasnodar at a depth of 14 m in the sandy loams of the second 
alluvial terrace of the Kuban. The death and deposition of the animals have 
been attributed to mudflow by Borisyak. 

Girei Quarry. The stratigraphy of the alluvial sediments of the 
Kuban Plain has been worked out by Gromov (1948) from the sections in 
the quarries near the village 'of Girei and the Kavkazskaya railroad station 
in the town of Kropotkin. 

Sandy loams and sands 2 m thick, and greenish loam 1.5 m thick 
with vertical plant roots have been exposed by quarrying in the 10— 15-m- 
thick Wurm terrace of the Kuban. Crossbedded gravels with sand lenses 
and gravel beds up to 3.5 m thick were exposed in the lower part of the 
quarry. Bones of mammoth occur at the base of the 3.5-m-thick sequence, 
which is underlain by a clay bed 0.8 m thick with plant remains (Figure 43), 
and a sand bed 3 m thick. The sand is underlain by gravels containing 
bones of bison, deer, and elephant. 

92 The following species occur in the gravel beds (Gromov, 1948): 
Elephas wusti, E. aff. trogontherii, E. primigenius. 
Rhinoceros mercki, R. an t i qu i t a t i s , Bison priscus cf. 
longicornis, B. priscus deminutus*. 

According to the observations of the geologist Kolbutov, the bones of 
E. wiisti and R. mercki came from the lower gravel bed, which 
overlies the clays with lignitized wood. The age of these fossils is close 
to those of the Tiraspol' gravel. The mammoths from the sands and gravels 
higher in the section are of Riss age, and thus contemporaneous with the 
"fauna" at the Il'skaya paleolithic locality (Gromov, 1948). The material 
represents three faunal complexes: l) post-Khazar with Elephas 
primigenius, 2) Khazar with E. trogontherii, 3) pre -Khazar with 
E. wusti. 

According to Gromov, the age of most of the bones from the Girei quarry 
is Khazar, i.e.. Early Riss or Riss-Mindel. 

Our studies in 1952 of the locality at Girei revealed the following 
stratigraphic sequence from top to bottom: surface loams; ancient alluvium, 
12-13 m thick; Sarmatian blue clay, 5-6 m of which were exposed by the 
Recent Kuban River erosion (Figure 44). 

Fragments of large tree trunks, often coated with ferruginous silt, 
saturated with water and charred in appearance, are common in the lower 
part of the sands and clays at a depth of 10 m. The following trees, which 

93 are characteristic of the present foothills, have been identified: oak 
(Quercus sp. ), beech (Fagus sp. ), and poplar (Populus sp. ). 
Bones occur in the gravels overlying the wood-bearing sands. Most of 
the bones are well-rounded, heavily permineralized and iron-stained. 



87 



(92) 




FIGURE 43. Gravels and sands (stripped of loam surface) in the Girei quarry 

Photograph by author, 1952 

We identified elephant teeth from the Girei quarry, now in the 
Krasnodar Museum, as follows: Elephas meridionalis — two molars 
with enamel 3.5-4 mm thick; E. trogontherii — three molars with 
enamel 2.4-2.8 mm thick; E. trogontherii primigenius — one tooth 
with enamel 1.8-2.2 mm thick. The bone material in the teeth of the 
meridionalis and trogontherii species is pink-tinged beige and 
resembles marl. The state of preservation is comparable in both species. 

3 

A tooth of an early mammoth with 8.5 enamel lobes over 10 cm on M is in 
a better state of preservation: in fracture the bone material is white, black- 
veined and slightly iron-stained. The collections also include bone fragments 
of deer (Cervus elaphus) and a marl-like fragment of a horn stem of 
Bison cf. s с h о e t e n s a с к i. 

The sandy-gravelly sequence at Kavkazskaya contains an assemblage 
consisting of animal remains which have been eroded and redeposited by 
lake sediments of different ages. As a whole, the Elephas meridiona- 
lis — E. trongotherii complex is possibly somewhat younger than the 
Taman faunal complex (see Chapter I). In all probability, there were no 
antelope in the younger complex: the bison were larger than the Taman 
species, and the deer were less diversified than those in the earlier complex. 

The bones of mammals which are infrequently found in the upper parts of the 
loess and in the fossiliferous soils underlying the loess on the water divides 
of northwestern Ciscaucasia are attributed to a later time. The Middle 
Pliocene localities in the lake sediments near Krylovskaya and Skobelevskaya 
are exceptions to this dating. Mirchink (l936b) has correlated the upper 



• The material was lost during the war. 




piuti^f, [riKil'.^Jfj^^-i^^^^^':^ 



вп 



-'-'-**i'-***-''»-^l'*l'-"'** ' 



ttm 



part of the loess in Ciscaucasia with the 
Wurm glaciation, and the fossil soils with 
the Riss-Wiirm interglacial. 

The Basal Quaternary of the Trans-Kuban 
Plain is represented by the middle part of 
the Psekups gravels with E. meridiona- 
lis, Equus stenonis and unidentified 
Bos (Gromov, 1948). 
94 Ц1^ Л 1,111111 1 iimif |ii 11-11Щ The meridionalis and trogonthe- 

rii elephant species are usually dated as 
the Lower or Middle Pleistocene. These 
occur, as a rule, in the highest terraces 
of the left tributaries of the Kuban. A tooth 
from the mandible ofElephas trogon- 
therii was found near the village of 
Pregradnaya on the Bolshoi Zelenchuk (ZIN 
collection). 

The Trans -Kuban Plain collections in 
the Krasnodar Museum include elephant 
molars of a species intermediate between 
Elephas trogontherii and 
E. primigenius. The tooth enamel 
is 1.8-2.4 mm thick; it forms 7-8 lobes over 
10 cm of the upper surface. The iron stain 
on the bones indicates that the remains of 
the elephants were buried in fine-grained 
river and lake sands, which subsequently 
became marshland. As for their state of 
preservation, these teeth always look younger 
than the teeth of E. meridionalis from 
the Psekups and Taman; the dentine and 
cement are dense and light beige and 
*\ [ I yellowish in color. In the gravel quarry 

I ^^^^^^^^^^^^^= near the village of Nekrasovskaya on the 

Laba, the following forms have been 
identified: tooth of an elephant of the type 
intermediate between E. na e r i d i о n a 1 i s 
and E. trogontherii; lower molar of 
Equus aff. sussenbornensis; 
fragment of a horn of the deer, Euclado- 
c e r u s sp. ; and a fragment of a horn of 
a giant deer, Megaceros euryceros. 
The faunal composition, the mode of 
preservation and the iron stain on the bones 
at this locality resemble some deposits on 
the Taman Peninsula. 
More recent collections by the Krasnodar Museum from the gravel 
quarry near the village of Ivanovskaya include molars of true mammoths 
with enamel 1.4-1.8 mm thick and 12 lobes over 10 cm. A molar of a 
mammoth was also found on the Pshish River . Teeth of Elephas 
meridionalis and E. trogontherii from the vicinity of the village 
of Novo-Labinskaya on the Laba River are in the Maikop Museum collection. 




1.5 



FIGURE 44. Section through bone- 
bearing gravels and sands in the Girei 
quarry 

a — surface loams; b — sandy loam with 
gravelly intercalations; с — river sands 
with layers of gravels and clays; d — 
gravels and sands with mammalian bones 
and tree trunks; e — Sarmatian blue 
clays. On the right — thickness in meters 



89 



The museum material is difficult to place stratigraphically because of 
inadequate labelling and the peculiarities of the mode of deposition on the 
piedmont plain (see above). On the plain, cones of silt debris, formed out 
of mudflow and carrying the bodies of animals, were deposited one on top 
of the other. However, these deposits were later eroded by rivers which 
redistributed the bones in aggradation terraces. In this way, faunas of 
quite different ages occur in the same river terraces, and, for this reason, 
Mirchink's (l936b) attempted synthesis of the geomorphological and paleon- 
tological data on the river terraces of the southern Russian Plain and of 
Ciscaucasia can be accepted only with great reservation. On the basis of 
the occurrence of Elephas planifrons teeth, Mirchink dated the 
175 m terrace near Cherkessk Pliocene in age. The 100-125 m terrace 
belongs to the Mindel glaciation period, according to Reingard. On the 
Kuban Plain, this terrace is overlain by two loess strata. Two 
layers of osseous soil, correlated with the "Paludina sands" near 
Taganrog, Bessergenovka and Yeisk, occur here. The 50m 
terrace near Cherkessk has been correlated by Mirchink, without 
paleontological evidence, with the moraines near the village of Khumarinskoe 
and the lower loess horizon occurring on the water divides of western 
Ciscaucasia. The loess, according to Mirchink, is of Riss age. It is likely 
that the loess in Ciscaucasia correlates with the gravels at the base of the 
95 second terrace in the lower Volga. Fossils of the Khazar faunal complex 
occur in these gravels. 

Il'skaya Paleolithic site. The activities of Paleolithic man in 
western Ciscaucasia are traceable through flint tools which occur on the 
shoals of the Kurdzhips and Belaya rivers near Maikop and Fortep'yanovka. 
The area probably had many summer encampments. The caves in the 
foothills, mentioned in local accounts, have not as yet been investigated 
by paleontologists. The Fanagoriiskaya cave, a narrow, cold crevice cut 
through the limestone conglomerates by the upper Ayuk rivulet west of the 
village of Goryachi Klyuch, was unsuitable for large animals. 

The Il'skaya site was discovered in 1898 by Debaille (see Volkov, 1899). 
This is an open site on the right bank of the 1Г, the left tributary of the 
Kuban, 40 km southwest of Krasnodar. The site, now in deforested foothills, 
was rediscovered in 1925 by Zamyatnin (1934) who carried out excavations 
in 1927-1928, and who was followed by Gorodtsov (1940) in 1930 and 1936- 
1937. 

Paleolithic remains occur in the lower of two beds of fossil soils in 
diluvial loams located on the second 10-14 m terrace of the IV under the 
dolomite cliffs. 

The archaeologists mentioned above have described numerous tools found 
at the site made of dolomite and, occasionally, of flint. Because of the 
poor finish of the dolomite tools, Zamyatnin dated the site Mousterian, 
but the more refined finish of the flint tools led Gorodtsov to date the site 
Solutrean. The geological dating of the terraces is Riss-Wurm, Early Riss 
(Gromov, 1933, 1948). 

According to Gorodtsov, the community abandoned the site when the 
surface of the Riss terrace became bogged at the onset of diluvial processes. 
At the same time oil seepage from the dolomite displaced the water at 
the site and preserved the strata containing Paleolithic tools and bones. 
The state of preservation of the bones varies at the Il'skaya site. Some 



90 



Э6 



fragments "jingle" and are so thoroughly permineralized that they only 
stick slightly to wet fingers; their specific gravity is almost equal to that of 
Pliocene bones. These bones are usually strongly stained with the iron 
oxide which fills the pores, or are covered with particles of silty fine-grained 
sand of a type characteristic of swampy springs on river floodplains . The 
enamel of the teeth is completely black. Those bones which were preserved 
in oil are younger in appearance. All the collected bones are food remains — 
the epiphyses on the bones are broken off, and the diaphyses split as though 
bone marrow has been extracted. Fragments of jaws and individual teeth 
of bison are particularly abundant and bones of young mammoths, showing 
teeth impressions of large carnivores, also occur at the site. 

When the asphalt which adhered to the teeth of the bison was dissolved 
with benzine, the remains of birch mice, small snakes, beetles and grasses 
became visible. 

Table 7 and Figure 45 give the species and their number in the Zamyatnin 
and Gorodtsov collections, as identified by Gromova (1932a, 1937) and by 
the author. 



TABLE 7. Mammals from the Il'skaya site 



Species and higher taxa 


Number of bones 


Number of 
individuals 


Ursus spelaeus 


2 
19 

1 

24 

210 

15 

2 

9 
13 
23 

1 

33 

37 

2,401 

38 

549 


2 


С an is aff.. lupus (small form) 

C. lupus 


4 
1 


Crocuta spelaea 


7 


Elephas primigenius 


5 


Lepus aff. europaeus 


1 


Sicista cf. caucasica 


2 


Muridae, not determined below generic level .... 
Equus caballus 


2 
5 


E. cf. hidruntinus 


3 


Sus scrofa 


1 


Cervus elaphus 


4 


Megaceros eurycercs 


4 


Bison priscus 


43 


Saiga tatarica 


2 


Artiodactyla, not determined below generic level 




Total .... 


3,377 


86 



The absence of woolly rhinoceros, elk and mountain goat is probably 
not accidental. These animals were rare and difficult to hunt. 

Elasmotherium and camel, common in the Middle Pleistocene of 
the Volga River region, either did not exist on the Trans -Kuban Plain, or, 
at most, were extremely rare. 

Bison remains account for 87% of the bones at the Il'skaya site. The 
species, according to Gromova, is one of the forms showing reduction in 
size: Bison priscus deminutus. 



91 



In fact, relatively small molars and carpal bones (of cows?) occur along 
with bones of large bison, which are indistinguishable from those of Khazar 
age of the middle and lower Volga. The proportion of large bones is small. 



(98) 




I -L 1 ' 



FIGURE 46. Mammal remains from the Il'skava site 



1— jawofCanis lupus; 2 — jaw of С rocu t a spelaea; 3, 4 — M4 of E 1 e p h a s pnmigenius 
showing worn surfaces; 5 — upper jaw of S i с is t a cf. с a u ca s i ca (x 8); 6 —phalange of E q u us 
hidruntinus; 7 — metacarpus of E. ca ba llus; 



1704 



92 



97 



but the determinant here may have been the selectivity of the ancient Trans - 
Kuban hunters rather than any morphogenetic factor. It is possible that 
hunters might have killed cows in preference to bison. 

The surroundings of the site were probably meadow- and forest-steppes, 
which were inhabited by large herds of bison, horse and deer. 

The climate and landscape of the Trans -Kuban region in the Upper 
Pleistocene can be best inferred from the presence of saiga and boar. Saiga 
did not live on the left bank of the Kuban in historical time, because of the 
damp summers and snowy winters which prevailed there. Boar, however, 
survived until recently by descending to the forested plain after heavy 
snowfalls in the mountains. Thus it can be seen from the evidence at the 
Paleolithic site that the Trans -Kuban area was more of a steppe than it is at 
present; the climate was drier, and there were no catastrophic snowfalls. 
However, the saigas could migrate from the north into the foothills only 
over ice- and snow-covered ground, as they do now in eastern Ciscaucasia. 




FIGURE 45 (continued) 

8 — jaw of Me gaceros euryceros; 9 — metacarpus of Cer vus ela ph us; 10 — horn stem of 
Saiga tatarica; 11, 12 — Мз and epiphysis of shank bone of Bison priscus^ 



93 



The negligible number of boar bones is probably accounted for by the 
much higher altitude of the mountain forests at that time. The absence of 
rock and mountain animals — mountain goats, sheep and boreal mammals — 
is also highly indicative. 

The absence of mountain goats, however, cannot be used in dating this 
site and estimating the extent of glaciation in the Caucasus. Even during 
the maximum extent of glaciation, these animals could have lived along the 
margins of the mountain glaciers near the Semashkho and Oplenen peaks, 
60-70 km from the site, and well outside the regular hunting grounds of the 
Il'skaya hunters. The mountains close to the site were low and gently sloped. 
The absence of the arctic fox, reindeer and ox indicates, most likely, that 
these species did not reach the northern Caucasus. 

The elephant bones at the Il'skaya site belong to a mammoth of a late 
type (tooth enamel thickness: 1.3-1.4 mm). At least two adults and three 
young, at the stage of tooth growth — M -M , were found. 

New collections of the bones of smaller species, which are preserved 
at the site in asphalt crusts, would contribute much to the understanding 
of the evolution of the Quaternary fauna in the Caucasus. 

Fragments of insect chitin which fell into the oil pools at the site and 
were thus preserved, were examined at our request by Bogachev, who 
identified eighteen beetles, one ant and one wasp (?). Of the beetles, the 
most common are water scavengers and water beetles (D у t i s с u s sp. 
nov. , Agabus sp. , Illibius sp. , Hydroporus sp. , Colymbetes 
f u s с u s L. ); others are ground beetles (Carabus sp. no v. , Platysma 
sp. , Amara sp. , Bembidium sp. ), leaf beetles (G a s t r о i d e a sp. , 
Donacia sp. ) and Coprinae (Aphodius subterraneus L. ). The 
steppe forms are represented by Tentyria nomas Pall. Many of the 
species differ considerably from the Recent forms, which indicates the 
antiquity of this fauna. 

Among the 20 grasses preserved in chunks of bitumen, collected by 
Zamyatnin and studied by Kipiani, the following predominated: Cruciferae 
gen. et sp. , Compos it ae gen. etsp., Lepidium perfoliatum, 
Polygonumi aviculare. Euphorbia cf. palustris, cf. Statice, 
Chenopodium sp. et cf. Atriplex. There were also occasional remains 
of Alisma sp., cf. Carex, Scirpus sp., Plantago sp., and other 
forms. No macroscopic remains of arboreal species were recorded. The 
species listed are ancient, weedlike types, characteristic of sunny habitats 
and some of them of a shallow -water environment. Cherdyntsev has set 
the age of the Il'skaya site from the carbon-dating of the bones at 39,000 
years . 

In our opinion, the Il'skaya fauna indicates the beginning of the Upper 
Pleistocene and tne extensive development of steppes in the foothills region. 

In 1957 Formozov carried out exploratory excavations at the entrance 
to a cave in the Belaya River ravine near the village of Dakhovskaya. He 
collected eighty fragments of poorly preserved bones and some tools of 
the Mousterian type from the 3-m level. We have identified the bones as 
follows: Canis sp. (10/2), Vulpes sp. (17/4), U r s u s s p e 1 ae us (3/2), 
Crocuta spelaea (l/l), Meles meles (7/2), Panthera spelaea 
(4/l), Felis cf. silvestris (2/l), Lepus europaeus (3/l), 
Cricetus cricetus (3/2), Elephas cf. primigenius (l4/l), 
Cervus elaphus (5/2), Megaceros euryceros (5/2), Capra cf. 



94 



caucasica (2/l), Bison priscus (19/3).- This fauna is characteristic 
of ecological assemblages of the Middle -to-Upper Pleistocene transition. 
As a whole, it is quite similar to the Il'skaya assemblage. Some small 
forms of wolf and fox are particularly interesting. 



Central Ciscaucasia 

On the Stavropol Plateau and in the Pyatigor'e area Pleistocene mammals 
occur mainly in diluvial loams covering the water divides and filling old 
100 gulleys. Older, Lower Pleistocene fossils occur in deposits of ancient 

streams and marine and lake bays on the surrounding plains. The Lower 
Pleistocene material is represented by Elasmotherium sibiricum 
from the vicinity of Divnoye in the Manych depression(Pavlova, 1916), and 
by Equus cf. stenonis teeth from loess loams near the village of 
Mirnyi (Khomenko, 1913a). The following finds can be dated Middle 
Pleistocene, Khazar: horn of Megaceros hibernicus, found by 
Khomenko in diluvial clays in the Kalaus cliffs; Saiga prisca 
(=S. tatarica) from diluvial clays in the Koronkoyar gulley, 35 km south- 
southwest of Stavropol; Equus caballus fossilis, in diluvial clays 
of the Shatyr mound, and remains of В о s sp. from sands at the bottom 
of a well in the stream bed of the Mutnyanka. 

The collection of elephant teeth in the Stavropol Museum includes three 
lower molars (7 lobes over 10 cm; enamel thickness 2.5-2.8 mm) of Lower 
Pleistocene Elephas t r о go nt he r i i. The museum collections also 
contain six fragments of long-horned bison skull and one fragment of skull 
of a primitive bull (Bos primigenius) which was found in the Stavropol 
area. Most of the material was collected from beds underlying the loess 
loams and can be dated Middle Pleistocene. 

New materials were collected by Gnilovskii in 1941 on the Kuban- 
Yegorlyk canal on the western edge of the Stavropol Plateau. The following 
specimens were collected from a 6-nn depth of the first terrace of the 
Yegorlyk: two fragments of mandible, a lower carpal epiphysis and a heel 
bone (bitten through by some carnivore) of Paracamelus cf. gigas; 
fragmentary bones of short -horned bison; and fragments of neck vertebrae, 
horn, forelimb bones and astragalus of Bison cf. schoetensacki. 
The bones are chalky in appearance, with light iron oxide stains along the 
cracks. 

The following bones were collected from a 9-mi depth in the alluvial bed 
overlying the Maikop clays in the first Yegorlyk terrace: fragments of upper 
molars of bulls (Bos sp. or Bison sp. ) and an upper molar and first 
phalanx of a small horse (Equus sp. ). The bones, which are iron-stained 
and not rounded, occur on the erosional surface and are probably Lower 
or Middle Pleistocene. 

Ryabinin's (1918) collections from the Udel'naya steppe of the upper 
Kalaus are of Upper Pleistocene and partly Holocene age. The material 
was collected from the upper part of the loess loams exposed in the left 
bank of the Malyi Kiyankul' gulley, at the upper end of the Pervaya 
Ternovaya gulley, and in the Kalaus stream bed near the opening of the 

* The numerator indicates the number of fragments; the denominator, a possible number of individuals. 



95 



Malyi Yankul' gulley. At the Malyi Yankul' locality numerous mammal bones 
occur at a depth of 2.5 m in the structureless brown, sandy, gypsiferous clay 
with small, limey concretions. Ryabinin has identified the following forms: 
"Equus caballus fossilis, Sus scrofa fossilis. Bos taurus 
fossilis, Saiga tatarica fossilis and a horn similar to those of 
goitered gazelle (Gaze 11a s u b gut t u r о s a). " Mesolithic tools (of the 
Solutreanor Magdalenian cultures, " according to Ryabinin) collected at 
the locality indicate the last part of the Pleistocene or even the transition 
to the Holocene. 
101 Thus, the fauna of the margins of the Stavropol Plateau does not support 

the hypothesis that mountain mammals of the Caucasus migrated to the 
plains during the maximum glacial advance. No boreal species have been 
recorded in the area. Studies of the caves on the southern and western 
slopes of Mount Strizhament (Vereshchagin and Gromov, 1953b) would 
contribute to the knowledge of the steppe and forest -steppe complexes of 
the Stavropol area of the Upper Pleistocene. 

Bones of a mammoth have been recorded from gravel quarries near 
Yessentuki in the region of the Pyatigor'e laccolith on the Kabarda sloping 
plain (Karpinskii and Dreier, 1904). Eikhwal'd (1850, p. 190) reported 
finding near Kislovodsk the complete jaws of a primitive form of Equus 
caballus covered with tuff. 

Travertines of the Pyatigor'e area. Travertines were 
deposited in post-Pliocene Pyatigor'e by mountain springs, which changed 
course after each mountain-building cycle. Animal remains were shifted 
and scattered by carnivores over the slopes until they became covered by 
sediment. Layers of lime -tuff formed on the bones upon contact with 
water rich in calcium-carbonate. 

Travertines were deposited on almost all the laccoliths — Lysaya, 
Zheleznaya, Zmeika and Beshtau. However, only the travertines of Mount 
Mashuk have been subjected to paleontological study (Ivanova, 1948; 
Mirchink, 1936b; Gromov, 1948). 

According to Ivanova, there are four stratigraphic complexes of 
travertines: the first of Pliocene age (see Chapter I), the other three of 
Pleistocene and Holocene age. 

The travertines of the second complex (Mindel and Mindel-Riss age) 
occur in patches on the northern and eastern slopes. Teeth and jaw 
fragments of E le ph a s aff. antiquus have been found in these 
travertines. The travertines of the third complex cover a larger area to 
the northeast. These travertines have been correlated with the 55-m Riss 
and Riss-Wurm terrace of the Podkumok, where numerous bones of deer 
(Cervidae) occur. The fragments of Lepus sp. and Capreolus sp. 
teeth, collected by Kobutov in the Boichenko quarry, probably come from 
the travertines of the third complex. The material is Upper Pleistocene 
in appearance. 

The travertines of the fourth (Upper Pleistocene to Recent) complex occur 
in the first and second terraces on the southern slope of a mountain near 
Pyatigorsk, which extends to the stream bed of the Podkumok. According 
to Gromov (1948), the teeth of an advanced type of Equus sp. and of 
Elephas primigenius found in the gravels and loams of the second 
(Wiirm) terrace originally came from the travertines. 



96 



The Pleistocene collections in the Pyatigorsk Museum include a sample 
of dolomitized travertine with a calcitic mold of a horn of a small bison 
(Bison sp. ). Also in the museum is a fragment of the upper jaw of a 
mammoth with the third milk tooth, collected from quarry No. 1 on the 
northern slope of Mount Mashuk. The fresh appearance of the bone is very 
unusual among fossil findings in the south and suggests that the Pyatigor'e 
mammoth died as recently as the Holocene. The fragment of mammoth 
upper molar found in the travertines near the Medovyi waterfall in the 
102 vicinity of Kislovodsk, the mammoth teeth found in the sands and loams 
near the village of Kamenskaya, and those found at a 12-m depth in the 
banks of the Yutsa River are all older in appearance. The museum 
collections also include two jaws of woolly rhinoceros (Rhinoceros 
tichorhinus) and six skull fragments of primitive bison. The state of 
preservation of this material is Upper Pleistocene. 

All these fossils from the alluvial and diluvial sands and loams of the 
sloping Central Ciscaucasian Plain give only a very schematic picture of 
the ecological assemblages of the mammoth fauna which lived near the 
mountains. In the Lower Pleistocene the region was undoubtedly a steppe. At 
the end of the Pleistocene, however, the region became forested, at least 
in Pyatigor'e as indicated by occurrences of deer and gazelle bones in 
the travertines of the third complex of Mount Mashuk, andby the occurrences 
of post -Pliocene "flora. " Krasnov has identified 14 species of grasses 
and trees from the travertines of the Pyatigorsk area — species which are 
also characteristic of Recent broadleaf forests of the foothills: beech 
(Fagus sylvatica), filbert (Corylius avellana), white beech 
(Carpinus betulus), elm (Ulmus campestris), oak (Quercus 
robur), Norway maple (A с e r platanoides) and other forms . The 
"flora" from Mount Zheleznaya, which is similar in composition, though 
somewhat later in age, was identified by Palibin (1913) as follows: willow 
(Salix alba and other species), linden (Tilia cordata), ash 
(Fraxinus e x с e 1 s i о r) and medlar (Mespilus germanica). No 
boreal elements have been recorded at these Upper Pleistocene or Lower 
Holocene localities; on the contrary, the southern elements are more 
abundant in the floras. 



Eastern Ciscaucasia 

Most of the plain in eastern Ciscaucasia was covered by the Caspian 
transgression, which accounts for the scarcity of Pleistocene mammals 
in the sediments of the plain. 

The semideserts and steppes in the lower reaches of the Kuma, Terek 
and Sulak were in the zone of subsidence. The postglacial marine and 
river sediments of the area are at present below the base of river erosion. 
As a consequence, the only known fossiliferous localities with Pleistocene 
mammals are situated farther west on the high terraces along the shores 
of Pleistocene bays, on the piedmont plain and in the foothills. 

The sloping plains of eastern Ciscaucasia are covered by diluvial 
sediments and the alluvium of the Terek, Sunzha, Sulak and other rivers. 
The relief of this area is more complex than that of western Ciscaucasia 



97 



as a result of uplift of the Terek and Sunzha ranges during the Tertiary and 
erosion and further peneplanation during the Pleistocere. These ridges 
now separate the Vladikavkaz [Ordzhonikidze] and Sunzha plains from the 
alluvial plains of the western Caspian region. 

As early as 1886, Pohlig mentioned mammoth bones from the Sunzha 
valley in the Caucasian Museum collection in Tbilisi. Mullaev (1929) 
described a tooth of E le ph a s trogontherii from the post -Tertiary 
beds on the Ursdon River near the village of Karagach in the southwestern 
corner of the Vladikavkaz Plain. Scapula fragments of E. trogontherii 
from the 5-m level of the Mozdok sand quarry are in the collections of the 
103 Pyatigorsk Museum. Teeth of E. trogontherii of an early type from 
the Grozny Region are in the collections of the Grozny Museum. 
All the fossils came primarily from Lower Quaternary alluvial sediments. 

Gromov (1948) recorded two skulls of Bison priscus de mi nut us 
and a fragment of a Megaceros sp. skull from the first terrace of the 
Terek near Mozdok. The cranial cavities of the bison skulls were filled 
with volcanic ash. This fact establishes the time of the fossils as the time 
of deposition of the volcanic ash which occurs under the gravels of the 
first terrace and on the face slope of the second Riss terrace on the 
eastern slope of Mount Mashuk. 

It seems possible that, in the last outburst of volcanic activity of the 
Kazbek and Elbrus in the Upper Pleistocene, animals were sometimes 
killed and buried in the ash deposited by torrents. Of the fossils from the 
Kuma valley below Budennovsk, a lower molar of a mammoth is included 
in the collections of the Pyatigorsk Museum. The Upper Pleistocene fossils 
are represented by fragments ofSus scrofa ferus mandibles collected 
by the geologist Plamenevskii in the alluvial sediments of the first (Wurm) 
terrace on the right bank of the Terek above Ordzhonikidze. 

Generally speaking, the evolution of landforms and faunal complexes 
was similar in eastern and western Ciscaucasia. The same large mammals 
occur in the Lower and Middle Pleistocene — Elephas trongotherii, 
giant deer, primitive bison and — in the Upper Pleistocene — boar. 

Forests were probably well developed in the Upper Pleistocene along 
the lower northern slopes of the Caucasus, and even in Dagestan. Whitebeam 
(Sorbus aria) and European hornbeam (Carpinus betulus)have 
been identified by Palibin (1913) as "post-Tertiary" species from the 
foothills near Makhachkala. These species also occur in the area at the 
present time. 

Very few fossil mammals are known from the piedmont and inner parts 
of Dagestan, or from the marine terraces of the western coast of the 
Caspian Sea, from Sulak to the Apsheron Peninsula. The collections in the 
museum at Makhachkala contain tooth fragments of Elephas trogon- 
therii of an early type (enamel thickness 2.5-3.0 mm) from the 
ferruginous gravels of the Manas River. A tooth fragment of a more 
recent elephant, with features of E le ph a s primigenius, and the atlas 
of an elephant from the vicinity of the village of Unty are in the same 
collections. 

Investigations of Paleolithic sites and excavations of caves and sheltered 
areas in the Apsheron limestone above Makhachkala and inner Dagestan 
would contribute to a better understanding of the evolution of Middle and 
Upper Pleistocene fauna in this part of the country. 



98 



Western Ciscaucasia 

In western Ciscaucasia the Pleistocene mammals occur mainly in cave 
deposits, kitchen middens and food remains of carnivores. 

Other types of deposition are seldom found. Several factors account for this: 
a lack of sedimentation and a rapid stream-flow which destroyed the bones, 
a humid climate which induced rapid decay of bone material, and the rapid 
104 subsidence of the sea bottom in coastal areas which precluded the formation 
of deltas . 

Some occurrences are known from terraces and from small reed and 
sphagnum marshes and lakes which still exist in the region of Sochi, 
Pitsunda, and Ochamchiri. Some of these localities were later uplifted and 
exposed by river erosion. There are extensive alder and sphagnum marshes 
along the lower Rion and Supsa (Poti, Kobuleti) in the area of the Colchis 
depression which is at present rapidly subsiding under the load of river 
deposits. Without artificial excavation, whatever fossil mammals occur in 
those beds will remain there for centuries to come, practically inaccessible 
to study. 

The tooth of Elephas primigenius Blum . in the Sochi Museum 
(mentioned by Gromov, 1948) might have been introduced into the area. 

Ryabinin (1937) has described a fragment of the right half of the mandible 
of a large cat, similar to Felis cf. spelaea, found in 1935 6 km north 
of Sochi. The mandible was taken from the gravels of the third marine 
terrace (Karangat) and originally dated Riss-Wiirm. More recently it 
was identified as Early Riss (Vardanyants, 1948). Shells of Cardium, 
Tapes and Mactra occur in the same terrace. 

The Black Sea coast of the Caucasus was favored by Paleolithic hunters 
for its propitious conditions for big-game hunting: numerous narrow valleys 
and canyons, coastal terraces which carried the free migration of ungulates 
both from the north and from the south, and many caves inhabited by bears. 

The uplift of the spurs of the Bolshoi Range in the Pleistocene saved 
the cave deposits from river erosion. Paleolithic sites occur along the 
foothills from Sochi and Tuapse in the north to Kutaisi and Batumi in the 
south and the east. Paleolithic deposits probably continue along the northern 
coast of Asia Minor. 

Zamyatnin (1937a) has described and mapped 43 Paleolithic localities in 
Abkhazia. He subdivided the sites into three chronological groups: l) pre- 
Mousterian, 2) Mousterian, 3) Upper Paleolithic. Most of the sites, 
however, are represented by only a few flint tools. 

Efimenko and Beregovaya (1941) have listed 15 open sites and 14 cave 
sites in western Georgia. 

Fossil mammals are known only from cave deposits in the gorges of the 
Khosta, Kudepsta, Mzymta, Tsebel'da, Rion, Kvirila and other smaller 
rivers. 



Caves on the Black Sea coast 

In 1936-1938 Zamyatnin (1940) excavated the Navalishinskaya and 
Akhshtyrskaya caves in the Sochi and Adler area in the Krasnodar Territory. 



99 



Navalishinskaya cave is near the village of Navalishino, 12 km 
from the sea, on the right bank of the Kudepsta. Two cultural layers were 
found in the cave: the upper layer (Upper Paleolithic) with bones of hamster 
and redeposited bones of cave bear; the lower layer (Mousterian, 
Aurignacian), with bones of elk, goat, hamster, cave bear, wolf, birds 
105 and shells of Anodonta and Helix. According to Zamyatnin, the cave 
was merely a seasonal shelter of early nnan. 

The mammals from the cave were identified by Gromov (1948), as shown 
in Table 8. 



TABLE 8. Species and number of bones of mammals from the Navalishinskaya cave 



Species 


Middle 
Paleolithic 


Upper 
Paleolithi : 


Canis lupus 


1 
165 

1 




Ursus spelaeus ,, 


229 


Cricetus cricetus 


1 


Alces alces 


1 


Caora SD 


1 






Total 


167 


232 







The Shirokopokosskaya cave (which Gromov called Navalishinskaya II) 
is located near Navalishinskaya I. It contained Upper Paleolithic tools, 
cave bear bones and a few bones of red deer (Gromov, 1948). 

Bones of cave and European brown bear were excavated in 1946 in the 
Vorontsovskaya cave in the Khosta ravine (collection of the Krasnodar 
Museum). The state of preservation of the cave bear bones is not uniform: 
in fresh fracture the mandible bone is brownish and free of organic matter. 
The radius and femur are dark brown, and show little loss of organic 
matter; when scraped, they exude an odor of fresh bones. It seems that 
the cave bear lived in postglacial times. 

Akhshtyr skaya cave is located on the right bank of the 
Mzymta River 100 m above the river bed at the point where the river leaves 
its narrow canyon, near the village of Akhshtyr, 15 km from Adler (Figure 
46), and is, at present, inhabited only by bats (mostly horsehoe bats). The 
entrance to the cave faces south and a corridor 4.35 m high, 2 m wide and 
over 100 m long leads to the cave, which was probably much lower in 
Paleolithic time. In front of the corridor is an overhead ledge and two 
entrances with two platforms. The eastern platform has been excavated 
by Zamyatnin (Figure 47). From the west the cave can be approached 
over a narrow, easily defended trail on the cliff. Zamyatnin assumed that 
it was intermittently inhabited by man over several millennia. Because of 
the elevation of the platform dwelling site, the inhabitants probably threw 
away most the bones and tools into the river. In the intervals 
between man's occupancy, bears and bats lived in the cave. Two narrow 
crevices, which were also inhabited by bears, are located farther west 
and down the cliff. 



100 



According to Gromov (1948) the big cave was formed during the formation 
of the 60-80 m terrace on the Black Sea coast, which was raised 112 m 
near the Akhshtyr anticline. 



106 




FIGURE 46. Mzymta gorge below the Akhshtyrskaya cave 

Photograph by author, 1954 

Two phases can be distinguished in the geological history of the cave. 
In the first phase two perpendicular faults formed: one, a normal fault, 
the other, a displacement along the bedding planes of the limestone. These 
two faults resulted in the formation of the cave which then, in the second 
phase, was continually enlarged by dissolution of the ceiling. During this 
period detritus, loam and fragments of stalactites continued to accumulate 
on the floor. 

Gromov assumed that detritus accumulated with cooling of the climate, 
whereas decomposition of the detritus into loam occurred in a warmer and 
more humid climate, but the faunal data which follow do not support these 
assumptions . 

Zamyatnin (1940) has identified five cultural layers in the cave from the 
Middle Paleolithic (Lower and Upper Mousterian) to the Recent. He 
correlated the Lower Mousterian of the Akhshtyrskaya cave with the beds at 
the Il'skaya site in Ciscaucasia and with the Kiik-Koba cave in the Crimea. 

The excavations produced over 3,000 identifiable fragments of bones, 
mostly from the Middle Paleolithic layers. The material includes bats, 
carnivores, rodents, ungulates and birds. 

Remains of large anim.als, fragments of epiphyses and diaphyses, are 
typical food remains. Bat bones and excreta have accumulated in the cave 



101 



into present tinne. Eagle owls probably carried in the hanister bones which 
were found in the cave. 

Most of the cave bear bones are metapodia, phalanges, isolated teeth and 
107 fragments of limb bones. Exploratory excavations in the sheltered section 
of the cave produced complete bones: femora, tibiae, fibulae, humeri, 
radii, ulnae, and skull fragments of bears which probably died of natural 
causes. 




FIGURE 47. Diagram of the frontal part of the Akhshtyrskaya cave 



The matrix on the bones of Mousterian and Upper Paleolithic specimens 
is indistinguishable to the eye, but specimens in varying states of 
preservation occur within the same bed. The bones are pinkish yellow to 
light brown in color. The enamel of the teeth is either the natural color 
or darker with black cracks, and the roots are yellow, brick or brown. 
The Mousterian bones are darker, with some surfaces spotted with dendrites; 
numerous dendrites may have coalesced resulting in a dark brown color 
over the entire bone. In fracture, the bones are usually light beige. They 
are light in weight and free of permineralization. 

The composition of the mammalian fauna of the cave is given in Table 9 
and Figure 48 (author's identification).* 

* Our count of bones and individuals in the ZIN collections differs somewhat from Gromova'S: bones from the 
Lower and Upper Mousterian were combined, since there is no noticeable difference in the composition, 
number and type of preservation. 

102 



108 A cervical vertebra of an eagle was found in the Mousterian beds. Bones 
of the goshawk Astur gentilis L. , of the magpie Pica pica L. and 
of small sparrows (Passeriformes) were collected from the Upper 
Paleolithic beds. 




FIGURE 48. Mammal remains from the Mousterian beds of the Akhshtyrskaya cave 

1 — metacaфal of С anis lupus; 2 — shankof Vu Ipes vulpes; 3, 4 — canine and astragalus 
ofUrsus spelaeus; 5 — femur of M a r t es sp. ; 6 — shank of С r i ce tu s cricetus; 
7 — carpal of S u s scrofa; 8 — second f)halange of С e г vus elaphus; 9 — first phalange 
ofCapreolus capreolus; 10 — carpal of A Ic es alces. 

The number of bones in the Upper Paleolithic beds is much smaller than 
in the Mousterian beds. The smallest possible number of cave bears is 39, 
as estimated by the number of phalanges. This was evidently a species 
easy to hunt in the canyons where it lived. The European brown bear is 
represented only by a molar from the Upper Paleolithic. 

Next in descending order of quantity are mountain goat, deer and boar. 

Remains of mountain and rock ungulates are fairly rare: the evidence 
of mountain goat has been mentioned, but there is no evidence of chamois. 



103 



109 The Upper Paleolithic shows a slight increase in the number of boars, 

which may indicate not only an increase in species population in the vicinity 
of the cave, but also the gradual extinction of the cave bear, which forced 
the hunters to look for other game. * 



TABLE 9. Mammals from the Akhshtyrskaya cave 





Middle Paleolithic 


Upper Pa 


leolithic 


Total 


Species 


number of 
bones 


number of 
individuals 


number of 
bones 


number of 
individuals 


individuals 


Rhinolophus fe rru m - e qu i n u m 
Vulpes vulpes 


9 
9 

2,946 
1 
1 
1 
2 

18 
2 

9 
6 


1 
1 

32 
1 

1 
1 

1 

3 
1 

2 

2 


1 
3 

1 

543 

1 

2 

7 
3 

4 

2 
9 


1 
1 

1 

7 
1 

1 

3 

1 
1 

1 

4 


1 
2 


Canis lupus 


1 




1 




39 


Martes cf. foina 


2 


Fells cf silvestris 


2 




1 


Sus scrofa 


4 


Alces alces 


1 


Cervus elaphus 


4 


Megaceros euryceros 

Capreolus capreolus 


1 

1 
2 


Capra caucasica 


6 






Total 


3,004 




576 




68 



The absence of wild sheep in the Upper Pleistocene of the Black Sea 
coast supports the notion that forests and mesophilous vegetation covered 
the area, as inferred from the composition of the Akhshtyrskaya "fauna. " 

It seems unusual that bison and giant deer bones were found only in 
Mousterian beds, since the bison has survived in this region to the present. 
Elk and gazelle appear only in the Upper Paleolithic beds, though these 
species existed in the Caucasus at least from the time of the Lower 
Pleistocene. Gromov's (1948, p. 262) theories of climatic changes which 
he predicated on the presence or absence of these species are, therefore, 
untenable. The number of specimens of ungulates and carnivores (except 
for the cave bear) is too small to furnish a basis for sound conclusions on 
faunal and climatic changes. 

The pale о geographic study of the animals from the Akhshtyrskaya cave 
must include the ecology of the species and the hunting traditions of the 
primitive tribes. 

The highly diversified terrain and the forests of the Black Sea coast 
were not suitable ground for mam:moths, horses, giant deer, saigas, 

• Gromova (1948) has identified a lower metatarsal epiphysis and the second phalanx of a ram as the small, 
wild "Ovis cf. ophion." Further examination shows that the bones are contemporary Holocene in 
appearance and their identification as Mousterian is incorrect; they are probably bones of a domestic 
sheep which fell into the excavation from post-Paleolithic beds. 



104 



110 primitive bison and steppe bison. The species occurring near the dwelling 
site were hunted for food. 

The topography of the Black Sea coast in the Pleistocene was such that 
the vertical zonation of the vegetation remained stable, despite climatic 
changes. This accounts for the homogeneity of the fauna in the 
Akhshtyrskaya cave throughout the Middle and Upper Paleolithic. 

The assemblages of mammals from the Navalishinskaya and 
Akhshtyrskaya caves are not universal. Nevertheless, they clearly indicate 
the existence in the western Caucasus of a boundary between the distribution 
areas of the steppe and forest-steppe species of the Russian Plain. Only 
the eurytopic hamsters crossed this barrier. No northern species are 
recorded from the western Caucasus and the northwestern foothills; nor 
were there any southern elements from the dry upland steppes of Asia Minor. 

Farther south along the Black Sea coast. Pleistocene mammals occur in 
the caves near Sukhumi (Krongel'm, 1873). They were recently discovered 
in the Ablaskira cave and in the grottoes in the Tsebel'da canyon. 

Ablaskira cave is located on the western slope of one of the spurs 
of the Panavskii (Kodor) ridge, 2 km northeast of the village of Otak and 
35 km from Ochamchiri. The Achkhitizgo River flows out of the cave 
entrance at an elevation of 26 5 m above sea level. 

Paleolithic remains occur in wall niches not far from the entrance. The 
cave has been described by Satunin (1910) and Maruashvili (1938b). 

Seventy-five fragments of bones of two adult cave bears and one young 
were collected in the cave by Sadovskii in 1946. The bones were collected in 
the section 50-400 m from the entrance in the clay and gravel of the stream. 
The material consisted of tibiae, humeri, metapodia, phalanges and a 
nearly complete cranium of a very short-snouted adult specimen. Like the 
animals found in the deeper part of the Akhshtyrskaya cave, these died from 
natural or accidental causes, rather than at the hand of man. In fracture, 
the bones are chalky white; they disintegrate easily under pressure, are not 
permineralized and their surface is free of dendrites. There are no data 
on the conditions of bone accumulation, the time of deposition of the clays 
on the cave floor or the time of human settlement in the cave. 

The grottos near the village of Tsebel'dinskoe in the Tsebel'da ravine 
were surveyed by the archaeologist Solov'ev in 1941. Paleolithic tools 
and bones collected in the Kholodnyi grotto and in Kei-Bogaz are now 
housed in the ZIN. The animal fossil material consists of fragments of 
teeth and of tubular and plate bones, which were transported partly by 
Paleolithic hunters and partly by eagle owls. Identification of the species 
in the collection is given in Table 10 and Figure 49. 

The state of preservation suggests that most of the bones are of Upper 
Pleistocene age; only a small part of the collection (the rodent bones 
in particular) may be of Holocene age. The bones represent the local 
faunal complex of mountains and forests. The southern limit of the common 
hamster is in the area of these southern caves. 



j^l]^ Caves of Imeretia 

The northeastern part of the Colchis Plain is a low plateau composed 
of limestones overlain by Oligocene sands and silts. The tributaries of 



105 



TABLE 10. Mammals from the Paleolithic sites at the Kholodnyi grotto and Kei-Bogaz 



Species 


Number 
of bones 


Species 


Number 
of bones 


Ursus spelaeus 


3 
1 
6 

1 

2 

20 


Sus scrofa 


75 


U.arctos 


Cervus elaphus 


60 


Martes sp 


Capra caucasica 


58 


Felis silvestris 


Bison priscus 


8 








Arvicola terrestris 


Total 






234 














FIGURE 49. Mammal remains from the Kei-Bogaz grotto 

1 — humerus of Martes sp.; 2 — humerus of Felis silvestris; 3— humerus 
ofCricetus cricetus; 4 — femur of A rv i с о la terrestris; 5 — astragalus 
of Sus scrofa; 6 — second phlanx of С er V us elaphus; 7 — first phalanx 
of Rupicapra rupicapra; 8 — epiphysis of metacarpal of Capra caucasica 



112 



the Rion cut canyons 50-60 m deep through the plateau. Over several 
millennia numerous caves (dated Upper Paleolithic) were formed in the walls 
of these canyons, and were inhabited by prehistoric hunters, cave bears 
and eagle owls. 

The first discoveries of the Paleolithic in western Transcaucasia were 
made in the first half of the last century. The Swiss Favre collected stone 
tools in 1863 from "Jason's Grotto" in the Sakazhia cave near Kutaisi 
(Zamyatnin, 1957). 



106 




FIGURE 50. Entrance to the Sakazhia cave 

Photograph by author, 1954 

Burnt and split bones were collected by the engineer Bernatskii (1884) 
in one of the caves near the village of Rgani in the Chiaturi area. Some of 
the bones were sent to Rudolf Virchow in Berlin, and some were left in 
Tbilisi, where they were studied by Anuchin (1887). Bernatskii's collection 
consisted mostly of bones of cave bear. 

Zamyatnin has dated all the Imeretian cave sites investigated thus far as 
Upper Paleolithic. The earliest sites were uncovered by Krukovskii in the 
Chiaturi area: the Khergulis cave near the village of Vachevi and the 
overhang, Taro-Klde, nearthe village of Shukruti. The "3rd layer" in the 
Khergulis cave contained Mousterian and Upper Paleolithic flint tools, and 
split bones and teeth of cave bear (Ursus spelaeus), horse (Equus 
cab alius), Caucasian goat (Capra cf. caucasica) and bison (Bison 
of. pr i s с u s). 

Krukovskii' s exploratory excavation at the Taro cave produced Upper 
Paleolithic cutting tools, Mousterian points, teeth of cave bear, bones of 



107 



Asia Minor hamster (Mesocricetus aff. au r at u s) and teeth and split 
bones of Caucasian goat and bison. ■'- 

According to Zamyatnin, the middle group of the Paleolithic sites includes 
the following caves: Sakazhia, Uvarovo, Bnele, and Devis-Khvreli. 
The Sagvardzhile cave on the Dzevrul rivulet belongs to the same group, 
although only preliminary data have so far been published on the many 
cultural layers of this cave (Kiladze, 1953; Gzelishvili, 1953). 

Sakazhia cave is on the left wall of the ravine of the Tskhali-Tsiteli 
rivulet (the right tributary of the Kvirila), near the Motsameti Monastery, 
113 3 km from Kutaisi. The cave, which was formed near the surface of the 
faulted limestone plateau, is now approximately 300 m above sea level 
and 50 m above the present stream bed. The arched entrance is 5 m high, 
broadening at the top, and leading into a corridor which turns left 
(Figures 50, 51). 

Even though few of the earlier boxtrees and hornbeams remain, the 
setting of the cave against the cascading river and fallen rocks is still very 
picturesque. The cave dwellers could hunt their game on the marsh plain 
of the Rion and in the nearby mountains. 

Indications are that these caves were not inhabited by eagle owls or 
bears, except possibly, in the latter case, for very brief periods. 




FIGURE 51. Diagram of the Sakazhia cave 
The bones have been identified by the author from drawings by S.N. Zamyatnin. 



108 



The Sakazhia deposits are 2 m thick. The cave was excavated in 1914 
by Shmidt and Kozlovskii who dug in the straight corridor, and in 1936 by 
G.K. Nioradze. In 19 54 the author collected some bones and flint tools at 
the far end of the cave where some deposits still remain. 

The several thousand flint tools and fragments, which include many 
cutting edges and knifelike tools (Figure 52), establish the age of the culture 
as Solutrean (Nioradze, 1953; Zamyatnin, 1957). The collection includes 
many points and arrowheads made of the diaphyses of horse and bison 
114 metapodia. It seems strange that the dating of the thick cultural layer is 





-I I 




FIGURE 52. Sakazhia cave flint tools (from Zamyatnin, 1957) and bone tools (from 
Nioradze, 1953) 



109 



so narrow, since the cave was undoubtedly used as a habitation into post- 
Solutrean time. Bones of domestic animals which occur in the collection 
are probably derived from the upper layers. 

Upper Paleolithic food remains are represented by teeth, phalanges, 
jaw fragments and epiphyses and diaphyses of skeletal bones. There are 
115 472 specimens in Shmidt's collection, 130 in the author's and over 16,000 in 
Nioradze's. * 

Most of the bones, however, were not identifiable, because they were 
split into fragments. Single bones (heel bone and radius of Felis 
spelaea) bear the tooth impressions of large carnivores. The enveloping 
matrix suggests Upper Pleistocene age. The fragments are mostly gray 
with a slight yellow, free of dendrites and fossilized, but not permineralized 
The tooth enamel is a natural color with only occasional darkening. The 
canines of a European brown bear appear to be younger — of iiolocene age. 

Belyaeva (1940c) and Gromov (1948) published lists of 18 species 
identified from a few teeth and phalanges in the Nioradze collection. The 
results of the author's studies of both collections are given in Table 11. 



TABLE 11. Mammals from Sakazhia cave 



Species 


Number 
of bones 


Number of 
individuals 


Vulpes vulpes 


5 

13 

35 

11 

12 

1 

3 

2 

1 

1 

1 

1U2 

1 

76 

140 

6 

5 

1,488 

166 

8 


2 




3 


Ursus spelaeus 


5 




3 


Panthera spelaea 


3 


Felis of. lynx 


1 


Martes cf foina 


1 


Males meles 


1 


Chionomys loberti 


1 


Castor fiber** 


1 


I^lystrix cf hirsutirostris 


1 




8 


E. hidruntinus*" 


1 




7 




10 


Capreoluscapreolns 


2 


Alces alces 


1 


Bison priscus 


32 


Capra caucasicat 


25 
2 






Total 


2,07G 


110 







* Shmidt's collection is kept in Odessa, Nioradze's in Tbilisi. 

*" The data on occurrences of these two species are tai<en from Gromov and Belyaeva, since they do not 
appear in any available collections, 
t In this and the following tables of Imeretian deposits, the goat species is represented by the west Caucasian 
form, rather than the east Caucasian, contrary to erroneous identifications by Gromov (1948) and 
Gromova (1948). 



110 



In addition to the bones listed in the table, ungulates and carnivores 
of intermediate size (goat, boar, chamois, wolf) are represented by at 
least 8,800 fragments of diaphyses, and larger animals (bison, deer, 
118 horse, cave bear. Panther a spelaea Goldf.) by at least 6,000 fragments . 
A few fragments of human bone were also found in the cave. 

The diversity of the landscape was reflected in a highly variegated species 
composition: bison, goat, deer, horse, boar and cave bear. Because the 
cave was advantageously located between mountain and plain, the Sakazhian 
dwellers were able to hunt boar, elk and beaver on the Rion marsh plain, 
bison (their chief prey), deer and horse on the Imeretian Plateau, and goat 
and chamois in the rocks of the Rion ravine. 

In the Uvarovo and Baratashvili grottos in the same 
canyon, Shmidt collected nearly 30 teeth of Caucasian goat, 12 teeth of bison, 
12 bone fragments of cave bear, 2 deer teeth, 1 horse tooth, 1 fibular tarsal bone 
of a wolf, 1 fragment of the pelvis of a beaver and a number of Upper 
Paleolithic tools. The state of preservation of this material is similar to 
that of material found in the Sakazhia cave. 

The Bnele cave collection consists of a few bone fragments, identified 
by Smirnov (1923-1924), together with bones from the younger site at the 
Gvardzhilas cave. 

The De vis - Khvr eli cave is located 4 km from the Kharagouli station 
on the right slope of the Chkherimela rivulet ravine in the Cenomanian 
crystalline limestones. The cave entrance, which faces southwest, lies 
340 m above sea level and 80 m above the Chkherimela rivulet bed. In 1926 
Nioradze excavated a section of the cave 25 m long, 4.5 mwide and 6.5 m high, 
which revealed four layers: l) upper layer, with Recent animals; 2) brown 
layer, with animal remains; 3) cultural layer; 4) clayeybed. Nioradze (1933) 
has identified the site as Upper Aurignacian, whereas Zamyatnin dates it Upper 
Magdalenian. The bone collection consists of food remains and bones in 
eagle owl pellets. The species, identified by Gromova (l929) and 
M. V. Pavlova, are listed in Table 12. (The number of bones available to 
Pavlova is unknown. ) 

TABLE 12. Mammals from the Devis-Khvreli cave 



Species 


Number 
of bones 


Species 


Number 
of bones 


Canis lupus 


? 

? 
20 

? 
30 

1 


Capreolus capreolus 

Alces alces 


2 




1 




Bos or Bison 


? 


Mesocricetus cf. auratus .... 

S us s с ro f a 

Cervus elaphus 


Capra cf. caucasica 

Rupicapra rupicapra 


22 
3 







As a whole, the fauna resembles the fauna of the Sakazhia cave. The 
smaller number of species recorded is due to the smaller scale of 
excavation. It is unusual that no equine bones should have been recorded 
at a site located in the center of a dry plateau. Near the spurs of the Rachin 
119 ridge are sites belonging to intermediate and late faunal complexes which 
are described below. 



Ill 



116) 




FIGURE 53. Mammal remains from the Sakazhia cave 

1, 2 — canines of Vul pes vulpes; 3, 4 — canines of M e 1 es meles; 5,6 — jaw of 
Martes sp,; 7— PmaFelis lynx; 8 — canine of Felis silvestris; 9 — Pm4and 
10 — canine of Pa nth era spelaea; 11 — MjCanis lupus; 12, 13 — metacarpal 
and metatarsal of U rsus spelaeus; 14 — jaw (x 2) and 15 — first molar (x 10) of 
Microtus cf. roberti; 16, 17 — Pmj and its worn surface from Equus caballus; 
18 — MsCervus elaphus; 19 — M3 R up i с a p r a rupicapra; 20— МзСарга 
caucasica; 21— Ms Bison priscus; 



112 



(117) 




FIGURE 53 (continued) 

22 — first phalanx of Sus scrofa; 23,24- first phalanx and astragalus of Cervus 
elaphus; 25, 28-30 — horn, astragalus, metacarpal and metatarsal epiphyses of 
Bison priscus; 26, 27 — first phalanx and astragalus of С a p ra caucasica 



113 



In 1934 Zamyatnin (1937b) collected Upper Paleolithic tools from beneath 
limestone overhangs on the right bank of the Kvirila near the village of 
Mgvimevi in the Chiaturi area. In 1940 Kiladze (1944) continued excavating 
under one of the overhangs. 

The bones from Mgvimevi are highly fragmented and free of organic matrix 
and permineralization except for a calcitic crust found on some bones. 

The bones collected by Zamyatnin and Kiladze were identified by 
Gromova (1948) and later by the author. The list of species is given in 
Table 13. 



TABLE 13. Mammals from the vicinity of Mgvimevi" 



120 



Species 


Number of bones 


Number of 
individuals 


Canis lupus 


3 

1 

1 

3 

21 

57 

37 

3 

40 


2 


Ursus spelaeus 


1 


Panthera spelaea 


1 




1 


Equus caballus 


3 


Bison priscus 


4 






О V i s cf . ammon 


3 


Caprovinae (not identified more precisely) 




To ta 1 


166 


15 







* The numbers include material received by the ZIN osteological laboratory after the 1948 
publication. 

The game aninnals in order of importance were bison, goat and ram, the 
latter represented by a noteworthy species of large argali. Horses were 
also abundant. The presence of cave bear and Panthera spelaea 
indicates the age as Pleistocene, possibly Topmost Pleistocene. The lack 
of deer, boar and gazelle specimens is due to the limited scope of the 
excavation. 

The Gvardzhilas cave is one of the Upper Paleolithic sites. It was 
partly excavated in 1915-1916 by the engineer Krukovskii (1916), and in 
1953 by Kalandadze and Tushabramishvili (1955). 

The cave is located in the ravine of the Chirula rivulet, the right 
tributary of the Kvirila, near the village of Rgani. The wide oval entrance 
faces south; it is located on a small protruding platform nearly halfway 
up the right wall of the canyon. Several grottos are located somewhat 
higher on the opposite wall. The length of the cave is 30 m, the width at 
the entrance is 19.5 m and the height approximately 4 m (Figures 54, 55). 

According to Krukovskii, the calcitic ceiling of the cave collapsed three 
or four times and two ancient cultural layers were buried beneath the fallen 
rubble . 

The tools from this cave are mostly cutting and scraping edges made 
of pink chert. Many bone implements were also collected: points, darts, 
awls (Figure 56). From the flint tools, the site was dated as Top 
Paleolithic, Magdalenian-Azillian. 



114 




FIGURE 54. Gvardzhilas cave 

Photograph by S.N. Zamyatnin, 1926 



Krukovskii collected nearly 150 bone fragments and teeth of carnivores, 
ungulates and rodents; 14 species have been identified by Smirnov (1923- 
1924). 

The collections made by Kalandadze and Tushabramishvili contain over 
4,600 bones; the author's contains 78 bones. 

All the bones of large animals are typical food remains of Paleolithic 
hunters: broken epiphyses, diaphyses, skulls and teeth. A small number 
of bones of insectivores, rodents and birds are derived from eagle owl 
pellets. The bones appear very fresh with a yellowish beige color 
Ч 2]^ and nearly intact organic matrix. Much of the enamel has retained its 
natural color showing only an occasional rusty tint or darkened crack. 
When the bones are scraped, they smell of dampness. Many show tooth 
impressions of field mice. The state of preservation shows that the bones 
cannot be older than Top Pleistocene and they are more likely to be of 
Lower Holocene age. 



115 





^ Spring 
FIGURE 55. Diagram of the Gvardzhilas cave 



The list of identified species from all the collections is given in Table 14. 
The collections from the two layers are combined for purposes of the count, 
since there are no differences between their species compositions. 

Nearly 900 fragments of diaphyses from Tushabramishvili's collection can 
with difficulty be identified as large ungulates, bison and deer; of which not 
less than 550 belong to wild goats, chamois and roe deer. In addition, the 
collection contains 2,970 unidentifiable fragments of small ungulates and 
carnivores (Figure 57). 

The sizable collection of Gvardzhilas cave material clearly indicates 
a considerable transformation of the ecological composition of regional 
mammals which occurred there [in the Colchis] at the Pleistocene -Holocene 
boundary. Horse, sheep and, probably, porcupine vanished from the 
plateau and, of the southern steppe-plateau forms, only the Asia Minor 
hamster remained. Since there are no reasons to suppose that the 
tastes and hunting habits of the primitive tribes changed at this time, it can 
be assumed that the disappearance of these species resulted from the 
afforestation of the highland. 

Smirnov, Gromov (1948) and Pidoplichko (1951) regarded the Gvardzhilas 
cave mammals as a stratigraphically and ecologically uniform assemblage. 
These authors attributed the downward shift of the altitudinal zones to the 
advancing glaciers of the Greater Caucasus. They cited as support the 
fact that primitive hunters did not hunt goats and chamois further than a 
day's march away with a heavy load, i.e., 15-20km. And they also considered 
that Prometheomys s с h ap о s с h n i ко v i Sat. (a present inhabitant) 
of alpine meadows) could not have existed in the cave environs in a zone 
of xerophilous vegetation such as exists there today. 



116 



Our observations indicate faulty reasoning on their part. The distribution 
of game animals, in particular that of Caucasian goat, was wider than at 
present. And even now, suitable ground for wild goat is to be found in rocks 
•■■^^ 6 km from the cave along the upper Chirula. In April 19 54, the author 
observed one eagle owl which regularly flew each evening from its day 
shelter below the Gvardzhilas cave to the nearest subalpine meadows and 
the forests of the Rachin ridge. This owl probably preyed on hamsters in 
the dry cave environment and hunted Prometheomys in the subalpine 
meadows and forests. 

It is evident that a discussion of the Gvardzhilas cave assemblage need 
not consider the shift of zones nor the mosaic structure of the landscape; 
neither should it preclude the possibility of two ecologically different types 
of animals coexisting in the sanne landscape zone: the alpine and 
mesophilous Caucasian goat and Prometheomys and the xerophilous 
Asia Minor hamster. 

The finds of Prometheomys and hamster remains indicate a 
Pleistocene faunistic connection between the western part of the Greater 
Caucasus and the Lesser Caucasus along the Imeretian Plateau and the 
western slope of the Surami Range. 



TABLE 14. Mammals from the Gvardzhilas cave 



Species 




Number of 
individuals 



Talpa aff, caucasica 

Erinaceus aff. europaeus 

Vulpes vulpes 

Canis lupus 

Ursus arctos meridionalis 

U. arctos caucasicus 

U.spelaeus 

Gulo gulo 

Mustela nivalis caucasica 

Mesocricetus aff. auratus 

Prometheomys aff. schaposchnikovi 

Sus scrofa 

Capreolus capreolus 

Cervus elaphus 

Capra cf, caucasica 

Rupicapra rupicapra 

Bison bonasus (caucasicus?) 

Total 



1 
2 
1 
2 
2 
1 
2 
1 
1 
8 
3 
2 
1 
5 
13 
1 



The Upper Pleistocene faunal complex of northeastern Colchis is known 
by the remains of game animals and eagle owl victims from five cave sites. 
The complex has some features which are like and some which are unlike 
the faunal complex of the Black Sea coast. The similarities are in the 
occurrences of the mountain-forest ungulates and carnivores: wolf, fox. 



117 



123 



bear, European wildcat, marten, boar, roe deer, deer, elk and bison, and 
in the complete absence of the indicator species of the Eurasian plains, i.e. 
mammoth and primitive bull. 

As opposed to the Black Sea faunal complex, a number of migrants from 
the high steppes of southwest Asia are present in the Colchis complex. 
These include the Asia Minor hamster, porcupine, argali, horse and 
Equus hidruntinus. 

These species migrated from the south into western Transcaucasia. 
Their migration was facilitated by the presence of a karst relief (i.e. , dry 
soil) and, to a lesser degree, by the arid climatic phase which occurred 
some time in the Middle Pleistocene. 




FIGURE 56. Flint and bone tools from the Gvardzhilas cave (from Zamyatnin, 1957) 



Forest animals (boar, elk, roe deer) were abundant in the complex, 
occurring mainly on the marsh plain of Colchis. The European beaver 
was an ancient relict. 

The alpine species of the highlands were much more widely distributed 
than in present time. 

The Kodor and Tsebel'da ravines formed "ecological barriers" for some 
northern (European) species and for the highland species of southwest Asia. 
These two faunas converged in Upper Pleistocene time in the region of the 
barrier. 



118 



Thirty species of mammals of Upper Pleistocene age are known from 
Paleolithic sites and other localities in western Transcaucasia. The species 
are distributed as follows: 

Insectivora 2 Rodentia 7 

Chiroptera 1 Perissodactyla 3 

Carnivora 13 Artiodactyla 13 

Lagomorpha 1 

It is significant that mammoth. Bos primigenius and northern 
forms (e.g., reindeer, arctic fox, blue hare) are missing from the complex. 
125 And yet there are no warm-climate southern forms in the complex either. 
The main difference between the western Transcaucasian and the western 
Ciscaucasian complexes is the presence in the former of forest and 
mountain species. This indicates that the contemporary fauna and landforms 
were developed in western Transcaucasia by Middle Pleistocene time. 

The transition to the Holocene (postglacial) fauna in western 
Transcaucasia was probably more abrupt than the changes which occurred 
during the Middle and Upper Pleistocene. At the end of the Upper Pleistocene 
the following changes took place in the fauna: the extinction of cave bear, 
cave cat and cave hyena; the southern retreat of southwest Asian upland 
xerophilous species (hamster, porcupine, argali); and a decrease among 
local forest and mountain species (bison, goat, chamois, roe deer). 
However, the fauna which has survived into the present has preserved the 
features of a forest fauna. 

The climatic change in western Transcaucasia since the Middle 
Pleistocene probably evolved in different ways than in other regions because 
of the montane character of the local topography and of regional climatic 
peculiarities. 

The paleogeographic picture we now have will undoubtedly be augmented 
by further studies of the karst in Imeretia, Abkhazia and the Krasnodar 
Territory. 

A special search for Lower and Upper Pleistocene cave sites must also 
be undertaken. 



Central Transcaucasia 

The regions of the Surami water divide, the upper Rion, Bol'shaya 
Liakhva and the Gori depression are of great interest to paleontologists and 
zoogeographers as the probable migration route of Neogene and Quaternary 
mammals from southwest Asia. Pleistocene mammals occur in Paleolithic 
caves and in diluvial deposits. 

In 1952 Lyubin (1953) was the first to find Lower Paleolithic tools in the 
Bol'shaya Liakhva valley and at MetekhskayaProna near Dzhidzhoita, Morgo 
and Lashe-Balta, at an altitude of nearly 1,500 m. The tools from Lashe- 
Balta (hand axes, flakes, cores) are made of andesite and other quartzitic 
rocks. All the specimens had been washed out of a high terrace (probably 
the 60 m terrace) and rounded and redeposited by streams As in Abkhazia, 
the bones, which were probably originally deposited in the terraces together 
with the tools, were not preserved. The value of the paleontological material 
from the lower layers of the caves is therefore all the greater. 



119 



:i24; 




FIGURE 57, Mammal remains from the Gvadzhilas cave 

1— jawof(x2)Talpa caucasica; 2 — jaw of Gulo aff. gulo; 3 — femur (X 2) of 
Mesocricetus auratus; 4 — jaw of P rom e t h eo m у s s с h a pose h n i к о v i; 5 — 
jawofCervus elaphus; 6-8 — M3, astragalus and metacarpal of С a p г a caucasica 
9- 10 — M2 and first phalanx of Bison bonasus. 



120 



Caves of Mount С has ovali -Khokh , Paleolithic and younger beds 
in caves of the Kudaro region in South Ossetia, were discovered by Lyubin 
in 1955. The four most important caves were located in the right wall of 
the Dzhodzhora rivulet ravine above the villages of Kvaisi and Chasovali. 
The caves are in the dolomite, 1,700 m above sea level and 300 m above 
the riverbed (Figure 58). Much material was collected from the upper cave, 
Kudaro I, which is dry and slopes gently to the north from two entrances 
facing south and east (Figures 59, 60). The excavations in 1955, 1956 and 
1957 exposed clay beds, 2.5 m thick, with blocks of limestone. Iron-, 
126 Bronze-, Copper- and Neolithic- Age tools and pottery occur in the upper 
part of the beds. Middle and Lower Paleolithic tools occur in the lower 
part of the section. The flakes, hand axes and points are made of quartzite 
and are in part Mousterian and in part Acheulean. All the beds contain 




FIGURE 58. Mount Chasovali-Khokh (caves indicated by arrows) 

Photograph by author, 1957 

bones brought into the cave by men and owls. The Lower Paleolithic bones 
are mostly devoid of organic matter; they are light brown in color, and have 
dendritic surfaces. Some bones were rounded like pebbles in a small stream 
which flowed through the cave in wet seasons. Fragments of limb bones, 
teeth, phalanges and metapodia of carnivores and ungulates predominate. 
The material is difficult to identify to the species level. 

Upper Paleolithic tools and well-preserved bones were found at a depth 
of 3 m in excavating the clayey-lime beds in front of the entrance to the 
Kudaro III cave. Table 15 and Figure 61 show the many mammalian 
species which have been identified from these bones; up to 1958, 40 
identifications from the Pleistocene had been made. 



121 



Collections from the Paleolithic beds of Kudaro I include: 42 bones of 
birds, 3 bones of lizards (Lacerta sp. ), 43 bones of toads (Bufo sp. ), 
22 bones of frogs (Ran a sp. ) and over 30,000 bones of fishes, primarily 
of Salmonidae. * 

The abundance of fish bones in the Lower Paleolithic beds might indicate 
that the beds were deposited under water, possibly in the paleo-Dzhodzhora 
River, which might have carried the bodies of animals and abraded the 
bones . 



127 TABLE 15. Paleolithic mammals from Kudaro caves 



Kudaro I 

Lower and Middle 

Paleolithic 



Kudaro II 



Kudaro III 

Upper 
Paleolithic 



T a 1 p a sp 

Neomys sp 

Chiroptera, not determined below the generic level 

Macaca sp 

Canis cf. lupus 

С uo n sp 

Vulpes cf. vulpes 

Crocuta spelaea 

Ursus spelaeus 

Martes cf. foina 

Mustela cf. nivalis 

Meles cf. meles 

Gulo cf. gulo 

Panthera spelaea 

P.pardus 

Carnivora, not determined below the generic level 

Lepus cf. europaeus 

Marmota sp 

Hystrix cf. leucura 

Allactaga cf. williamsi 

Mesocricetus aff. auratus 

Microtus aff. majori 

M. aff. gud 

Prometheomys aff. sc ha poschni ко v i ... 
Muridae, not determined below the generic level 

Rhinoceros sp 

Sus scrofa 

Cervus cf. elaphus 

Capreolus aff. capreolus 

Capra cf. caucasica 

Ovis cf. ammon 

Ovis or Capra, not determined below the generic 

level 

Bison priscus 

Small bone fragments of ungulates and carnivores - 

Artiodactyla-Carnivora 



1 

1 
27 

4 
27 

3 
33 

1 
2.979 

9 

1 
11 
11 

3 

9 
18 
12 
10 
18 

2 
56 
16 

5 
13 
52 

6 

2 

213 

36 

55 

26 

189 
5 

13,013 



20 



135 



In a communication on the 1955 excavation (Vereshchagin, 1957b) the intermaxillary bone of a salmon 
was erroneously identified as a bone of a large lizard. 



122 



(128) 




FIGURE 59. Diagram of Kudaro I 
1,2— excavations of 1955- 1958 



As a whole, the Kudaro I fauna is indicative of dry plateaus and includes 
widely distributed forms (wolf, fox, cave bear, badger), Caucasian mountain 
forms (panther, marmot), and southwest Asian forms (hamster, sheep, 
porcupine). The indications of the migration of southwest Asian upland-steppe 
forms to the Caucasian mountains and of the survival of apes through the 



123 



end of the Pliocene are matters of particular interest. Porcupines, 
sheep, a few goats and chamois indicate that the relief was quite uniform 
and the climate warni and dry. 
128 Comparison of this site with the Upper Paleolithic sites of Imeretia 

suggests that mountain goat and other mountain forms existed in isolated 
areas of Caucasia which were undergoing differential uplift during the 
Pleistocene. In the Lower Paleolithic the upper reaches of the Rion were 
uplifted as much as 500 m. 

The Upper Pleistocene fauna of central Transcaucasia is represented 
by single mammoth teeth. One of the teeth of E le ph a s primigenius 
from the Dusheti area has been described by Burchak- Abramovich (1946). 
The tooth of a small mammoth was also found in the town of Gori in the 
first terrace of the Kura River; according to Gabuniya (1952a), it belongs 
to an Upper Pleistocene or Lower Holocene mammoth. It is probably of 
the same age as the late man^moths of the Pyatigor'e area. 

The presence of mammoths in the Upper Pleistocene of Transcaucasia 
is not proof of a migration of "northern" forms to the south, but rather 
is probably indicative of an endemic evolution of the proboscideans, such as 
occurred in the western Mediterranean. 



:i29] 




FIGURE 60. Excavations at southern entrance to Kudaro I 
Photograph by author, 1956 



1704 



124 



Eastern Transcaucasia 

In post-Tertiary time eastern Transcaucasia was plain country 
surrounded by mountains in the north, west and south, and the sea in the east . 

During the Pleistocene the seas transgressed at least three times over 
the eastern part of the plain, considerably reducing the living space of the 
land animals. 
129 The known localities of Quaternary mammals are on the marine terraces 
of the Apsheron Peninsula and on the foothill river terraces in the western 
part of the plain. Bone accumulations in caves are almost unknown in 
eastern Transcaucasia. 

Lower Pleistocene mammals occur in littoral sediments of the Baku sea 
on the Apsheron Peninsula. Skull fragments of a young rhinoceros 
(Rhinoceros cf. mercki), large horse (Equus sp. ) and hyena 
(Crocuta sp. ) were collected at a depth of 2 m in coarse, shelly 
limestone with Didacna rudis Nal. and D. surachanica Andr. near 
the Kishly railroad station, north of Baku. (The skull of the hyena was 
previously identified as Hyaena striata (Bogachev, 1925a, 1938b), 
which is impossible for the Lower Pleistocene. ) A fragment of mandible 
and a tooth of Elasmotherium from the 12-m level in No. 19 oil well 
near the village of Binagady are in the PIN collections. This material comes 
from an ancient Caspian terrace with Didacna surachanica Andr. 
The southern part of the terrace underlies kir-i= beds which contain 
the Binagady faunal complex (discussed below). 

A lower jaw of a colt (Bogachev, 1938c), identified by Gromova (1949) 
as Equus aff. siissenbornensis, was collected in oil-bearing 
gravels of the ancient Caspian terrace which overlies the oil-bearing beds 
at Khurdalan on the Apsheron Peninsula. 

Fossil mammals are more common in Middle and Upper Pleistocene beds 
than in the Lower Pleistocene. However, their dating is sometimes 
difficult because of the poor state of stratigraphic knowledge and the 
discontinuities of the Caspian terraces. Most of the material comes from 
the Apsheron Peninsula. 

The preservation of material in this region was promoted by considerable 
changes in the relief caused by tectonic, erosional and eolian activity. The 
ancient, transgressing and regressing Caspian waters in some cases obliterated 
131 and in others enhanced the work of the forces operating on the land. 
Following the major transgressions of the Caspian, cones of mud 
volcanos appeared on the Apsheron Peninsula. As a rule, all the deposition 
sites of post-Baku fossils occur in oxidized oil shows, i.e. , bitumens and 
asphalts. 

Many animals died in oil and liquid asphalt which preserved the bodies 
in a perfect state. The bituminous beds, because of their plasticity, 
protected the bones from erosion by water and wind. 

In most cases. Pleistocene fossiliferous sites were formed through the 
compound action of freshwater and oil or asphalt seeps on accumulations of 
mammal remains. 

The best known are the Binagady and Kir-Maku fossiliferous localities, 
although fossil mammals also occur at other places, including; 
a jaw of a camel (C ame lus sp. ) found in a ditch at Surakhany; skulls and 
horns of red deer (Cervus elaphus) collected in the bitumens on 
Artem Island (Bogachev, 1924, 1932); maxilla of a horse of the "Asian type" 

• Kir, or bitumen, was and still is used in covering and sealing flat roofs. 

125 



дзо) 




FIGURE 61. Vertebrates from the Paleolithic beds of Kudaro I 

1 -intermaxillary (X 2) of salmon Salmo sp. ; 2 - M^ (x 2) of M a с a ca sp. ; 3-5 - milk canine (X2X 
Mj and metatarsal of и rsus spelaeus; 6 — upper jaw of Pa n the ra pardus; 7 — jaw of M a rm о t a 
sp.; 8 — jaw (X 2)of M esocricetus aff. auratus; 9 — side and worn surface of tooth (x 2) of 
Hystrix sp. ; 10 — phalanx of Rhinoceros sp. ; 11 — horn ofCapreolus cf. capreolus; 
12— first phalanx of Cervus cf. elaphus; 13-15 — metacarpus, first phalanx and astragalus of 
Ovis cf, ammon 



126 



(Equus cf. caballus) found in the clayey breccia of Mount Bog-Boga in 
Balakhany, 15 km northeast of Baku; bones of a horse known from the ancient 
Caspian terrace on the Tertiary hills near the village of Ali-Bairamly; 
remains of one Bos , recorded from the Baba-Zanan Tertiary hills near 
Sal'yany (Bogachev, 1938c). 



Binagady 

The Binagady fossils are important in the study of the Quaternary fauna 
of the Caucasus. The occurrences at that locality of plants, mollusks, 
insects, reptiles, birds and animals of all ecological and morphological 
types allows a detailed reconstruction of the ancient Apsheron landscape. 

The locality was discovered in 1938 by a student, Mastan-Zade, who 
was studying the Apsheron bitumens. The first excavations were organized 
by Bogachev in 1938 (Azerbaidzhan branch of the Academy of Sciences of 
the U.S.S.R.). The excavations continued until 1941 with the participation 
of Kasabova and Sultanov and were resumed in 1946 under the supervision 
of Burchak-Abramovich. 

The primitive mining of kir over many centuries left holes and 
pits of various sizes in the bone -bearing area (Figure 62) which 
seriously interfered with the excavations. 

In order to obtain a clear stratigraphic picture the locality must be 
excavated in a series of mutually perpendicular trenches. 



(132) 




FIGURE 62, Binagady hollow remaining from bitumen-mining in area where rhinoceros bones were found 

Photograph by author, 1941 



127 



The Binagady locality is on the crest of a hill 0.5 km southeast of the 
village of Binagady, and 7 km north of Baku. The coast at its closest is 
10 km to the south and 25 km to the north. 

132 The bone-bearing area comprises approximately I2 hectares and is 
located on a hilltop near the Kyrrar hill. The area is 54-57 m above 
present sea level, and 48 m above the level of Lake Beyuk-Shor. An ancient 
mud volcano (Kichik-Dag) lies north of the fossiliferous area; further 

to the north is the meridionally elongated, saline Lake Masazyrskoe 
(Mirdalyabi) and to the northeast. Lake Binagady. To the east is the saline 
depression Kariatakh-Shor, beyond which rises the Balakhany Plateau. 
Extending from the Binagady hill are oil-bearing Salinas and the 
saline Lake Beyuk-Shor, which stretches far to the southeast (Figure 63). 

According to the Volarovich-Lednev map (1913-1929) and Bogachev's 
description (l940b), the base of the Binagady hill is a broad, anticlinal 
fold made up of Maikop (Oligocene) and diatomaceous (Miocene) clays and 
of productive sandstones (Middle Pliocene). South of the hill is the Middle 
Quaternary Caspian terrace. An eroded cone of the mud volcano Beyuk- 
Dag, 105 m above sea level, is of Pleistocene age. The volcano probably 
formed on the northeastern part of the Miocene fold in two stages. The 
loessial sands at its base are dated Qg^MRiss-Wurm) and at its top Q2'' (Early 
Wurm). 

The crest of the fold was eroded by waves of the transgressing sea. 
As a result of erosion the oil from the productive beds seeped into the 

133 coquinas of the ancient Caspian terrace and into the later Quaternary beds. 
The pressure within the productive beds resulted in oil and asphalt seeps 
at the surface of the fold and subsequent cementation of the overlying 
sediments by asphalt. Consequently, a series of fossiliferous asphalt 
layers of varying ages appears in the syncline. The most recent asphalt 
seep occurs on the southeastern slope of the hill; it probably formed during 
oil formation in the last two centuries. 




FIGURE 63. Present topography of the bone-bearing localities at Binagady and Kir-Maku 

The bituminous area extends to the ancient Caspian "Khasar" terrace 
in the south. The following fossils are characteristic of this terrace: 
Dreissensia polymorpha Pall., Didacna surachanica Andr. 



128 



D. kovalevskii V. Bog., Neritina pallasii Lindh. It is possible 
that the lower part of the bone-bearing beds correlates with the near-shore 
coquina of the Khazar terrace. The terrace slopes gently downward to the 
south, an indication of tectonic movements in post-Khazar time. It does 
not extend beyond the northern slope, which, in Khazar time, was probably 
dry land which was later subjected to strong erosion. If, as agreed upon 
by many geologists, the landscape of the area at the time when the bone- 
bearing beds were being formed was similar to the Recent landscape, 
it follows that during the marine transgressions the area became an 
archipelago of small islands and peninsulas. 

The present-day relief of the Apsheron Peninsula was formed through 
the interaction of tectonic forces and eolian and subaqueous erosion 
(M. Mirchink, 1934). The incidence of shallow, elongated salt lakes in the 
area is a result of wind erosion. Even the deep Yasamal'skaya valley, 
extending from Binagady to the sea, was probably formed by wind erosion. 
The Apsheron limestones which cover the Tertiary loams in places were 
evidently not sufficient to protect the loains from erosion by northerly winds. 
The Upper Pleistocene and Holocene produced tectonic and epeirogenic 
movements which were responsible for much important change in the relief 
of the Apsheron Peninsula. 

134 According to Shlepnev (1947), the epeirogenic movements in the 
Apsheron over a mere 16-year period (1912-1928) resulted in differences 
in elevation between various points as great as 114 cm. Thus the present- 
day relief in the Binagady area cannot be used as a starting point in the 
restoration of the relief at the time of bone deposition. 

A generalized stratigraphic section from the center of the bone-bearing 
area is given in Figure 64. Surface layer A is diluvium, 0.2 to 1.0 m thick, 
underlain by bedB, which is 0.75 to 1.4 m thick and contains fine-grained 
sand with fairly coarse coquina of marine moUusks, impregnated with oil. 
The southern portion contains large trunks and branches of juniper and 
occasional bones of horses, deer and large birds. Bed С consists of dense, 
viscous, dark brown bitumen, 0.1 to 0.5 m thick, which seems to form a 
lens toward the margins of the bone -bearing area. Pockets near the top 
of the bitumen bed contain bones of birds (mainly ducks), beetles and grasses. 
Bones of large animals are embedded in the bitumen of bed C, and 
occasionally at the boundary of beds В and C. 

BedD is fine-grained sand with fine coquina and some small pebbles; 
it measures 0.3 to 0.6 m in thickness. Horizontal bedding, produced by 
streams, and films of narrov/, swamp-plant leaves occur in this bed, which 
is the main bone-bearing bed. It contains bones of mammals, birds and 
reptiles, as well as insects, mollusks and plants. 

The granulometric analysis of beds BandC shows that the fine-grained 
material amounts to 62.5 and 75.4%. This probably indicates the littoral 
eolian origin of the Binagady beds (Vereshchagin, 1951b). 

Fossil animals and plants occur in pockets in the oil-impregnated sand 
which underlies- the bitumen bed. These pockets are more heavily 
impregnated with bitumen than the surrounding rock. 

The osseous strata 3 m from the surface gradually grades into the 
fine-grained sands of the productive beds, which are slightly impregnated 
with oil. A bitumien bed, 10-15 cm thick, can be seen in the holes dug at 

135 the highest point of the hill. This bed, 0.8-0.1 [?0. 8-1.0] m from the 



129 



A < 



D ■< 



E < 



II = II ■=■ II . 



=^==fL=^l 0.2-1.0 



ai-u 



ai'0.5 



03'0£ 



surface in Holocene loams, was probably 
formed during a recent phase in the 
migration of oil. It is rich in bones of small 
birds . 

The origin of deposition of the beds shown 
in the section is not entirely clear. Bed В 
was probably formed along the shore of a 
body of standing water, as indicated by the 
coarse coquina of marine shells composed 
of fragments 2-5 m in diameter. Bed С was 
probably formed as a result of migration of 
oil along a fault in the productive beds 
somewhat to the north of the Pleistocene 
bituminous lense. 

Mastan-Zade (1939) has hypothesized 
that an oil flow occurred during the eruption 
of the Kichik-Dag volcano. 

It is obvious that bitum.inization (oxidation 
and thickening of oil) occurred at the surface. 
This being the case, bed В must have been 
formed during a transgression in the basin 
which covered the bituminous lens, since 
the bituminization could not have occurred 
under water or else the oil floating to the 
surface would have been dispersed by wind. 

Bed D was probably formed in a short- 
lived body of stagnant water. It is difficult 
to detect the boundary between the osseous 
bed and the productive sands, because it 
is not clearly defined. The negligible 
percentage (0.2-1%) of shell fragments, 
2.0-3.5 mm in diameter, indicates littoral 
marine or estuarine deposits, in addition 
to windblown, fine-grained sand. 

The incidence of xerophilous grasses, 
small tree branches and detritus mixed with 
aninnal bones and fossil beetles indicates 
that the material was deposited in topographic 
depressions from small mud-oil flows. 
Individual bird bones, remains of insects 
and plants, skulls of animals filled with bituminous material — all these 
indicate poor sorting of the kind that occurs in a river flow. The likelihood 
is that the remains accumulated in a stagnant basin where the bones were only 
slightly moved over the substrate. 

With the exception of individual skeletons of small birds (dove-size or 
smaller) all the skeletons at the site are disjointed. Rennains of ligaments 
and feathers occur only on the bones of birds from the younger beds, e. g. , 
the bituminous intercalation in bedB. No complete skeletons of large 
animals were found in the first excavation, possibly due to the scale of the 
excavation. Scattered bones of large and medium-sized animals occur in 
the asphalt beds mostly in horizontal positions. Burchak-Abramovich has 
found bones of a horse and of a deer scattered over an area of several 




FIGURE 64. Section through the asphalt 
beds at Binagady 

A - surface loams; В — bituminous sands 
with coquina, bones and wood; C— bed 
of "rich" asphalt; D — oU-impregnaied 
sands with animal and plant remains; 
E — sands of productive beds. On the 
right — thickness in meters 



130 



square meters. In the bituminous bed complete skulls of medium-sized 
animals (wolf, hyena, badger) were found separately from the cervical 
vertebrae; even the atlas was separated from the skull. One find has been 
recorded which consisted of six joined cervical vertebrae of an ass and 
the mandible of a horse in two sections: the anterior part, which had been 
broken off, was deposited in a bed adjoining the horizontal layer where 
the posterior section lay. Broken bones of large animals and birds occur, 
as a rule, in the lower part of bed B. 

The deposits in bedD are characterized by quite complete bones, 
135 including some broken epiphyses and intact skeletal bones of large birds — 
all with asphalt -filled cavities as a result of the penetration of oil into the 
bone tissue pores. 

One seldom finds bones that show tooth impressions of wolves, bears 
or hyenas, although gnawed humeri of bears, the pelvis of Equus 
hemionus and the femur of a rhinoceros have been described by 
Vereshchagin (1951b) and by Burchak-Abramovich (1953a). There are no 
signs on the rodent skulls of beak or claw puncture by birds of prey, which 
indicates that the bitumen-preserved skulls probably were not deposited 
in the excreta or stomachs of birds which fell into the asphalt. Puparia of 
flesh flies occur, although not frequently, in nasal and brain cavities, and 
in the depressions of a rhinoceros humerus. When the puparia were opened, 
it could be observed that the flies had hatched from the pupae. There were 
no bones found which displayed cuts or breaks made by prehistoric man. 
Paleopathological evidence of osteomyelitis, rachitis and helminthiosis 
is common (Vereshchagin, 1946b, 1951b; G. V. and D. V. Gadzhiev, 
1952). Only two bones of a wolf and deer with polished distal epiphyses 
are known. Among the Binagady bones, there are only very few which have 
been rounded by streams, or which show sand-polished surfaces or other 
signs, such as brittleness or cracking, of origins in land surfaces or 
well-aerated soils. 

The extent of permineralization of the bones is relatively slight. The 
diaphyses, when they are sawed, smell of damp bone, which is unusual for 
Middle Pleistocene material. Proteins were detected in the humerus of a 
wolf, and juniper wood from bed В contained 30.52% cellulose (Mamedaliev 
and Kaplan, 1948). 

The dry bones and polished surfaces of diaphyses are dark brown and, 
in some places, bronze. In fresh fracture the bones show a grayish brown 
or chestnut color. Upon being boiled in alkali, the bones become brittle 
and light chestnut in color. The enamel and dentine of teeth are usually 
gray-brown to chestnut. However, the large molars of hyenas, saigas and 
bulls are bluish in color on the inner surface. 

Because the extensive excavations of 1938-1942 at Binagady were poorly 
documented, very little is known about the nature of the deposition. 

Almost all the investigators who worked with the Binagady material were 
concerned with the Apsheron landscape and the origins of the deposit: the 
reasons for death and the circumstances under which the remains accumulated. 

Bogachev (l939c, 1940b, 1944) maintains that the present-day relief 
is very old and that the contemporary Lake Beyuk-Shor is a relict of an 
ancient and much larger lake. He thought that rhinoceroces, horses. Bos 
and deer attempting to reach clear water were trapped in the oil-surfaced 
lake, and that quadruped and feathered predators attracted to the floating 
bodies fought and died over their prey. According to Bogachev, the region 



131 



was also inhabited by migratory birds, and numerous streams 
presumably entered the lake carrying plant detritus, juniper trunks and 
pear fruit. He assumes the climate of the region to be somewhat colder 

137 and more humid than that of today. 

Considering all the evidence, Bogachev compares the locality with the 
asphalt deposits at Starina (Galicia) and at La Brea (Los Angeles, California). 
He points out that organic remains accumulated gradually without any of 
those catastrophic episodes which kill thousands of animals of 
morphologically and ecologically different species. 

There are, however, a number of facts which do not fit into this elegant 
picture. If the relief in the past were similar to the present, then the 
Binagady hill would have been a small island when the level of the lake was 
higher and there would not have been room for all the animals nor for the 
accumulation of oil on such a small island. 

The absence of freshwater mollusks and fish in the bone -bearing bed 
and the concomitant incidence of xerophilous plants do not support a picture 
of a large freshwater lake. And, indeed, if there had been numerous clear 
streams, there seems little reason for animals to have waded into oily mud. 
The idea of rhinoceroses and boars dying trapped in mud can practically 
be dismissed. There are no silt deposits in the Binagady beds, and the 
animals could not sink in the very thin layer of fine-grained sand and 
coquina. 

From the plants identified in the asphalt, the botanist Petrov (1939) 
has inferred the existence of savannah and sparse arid forests in the 
Apsheron. He thought the site of deposit was in the delta region. 

According to Argiropulo and Bogachev (1939), the occurrences of forest 
dormouse, red-tailed gerbil, migratory (gray) hamster and jerboa, and 
of mesophilous species of beetles indicate that the forests were of tugai 
type and the climate was more humid than now. 

The ornithologist Serebrovskii (1948), in an analysis of the avifauna, 
included migratory birds in the species composition. He visualized the 
area of the locality as a shallow, reed-grown swamp surrounding a marine 
bay. Bird casualties were attributed by Serebrovskii to oil film on the 
water. 

According to the geologist Sultanov (1947), the deposit site was located 
on the shores of a shallow marine bay beyond the main zone of wave action. 
His conclusions that rhinoceroses, Ellobius, mole voles and jerboas 
drowned in the sea and that flying birds died of fumarole gas poisoning 
do not require comment. 

Efremov's statement (1950) in his excellent "Taphonomy" that bone- 
bearing bed D is alluvial in origin and was presumably protected by the 
asphalt from later erosion is in error. 

Gromov (1952), who studied the rodents, adopted Efremov's taphonomical 
viewpoint and explained the accumulation of organic remains at Binagady 
by the agency of river sediments. He assumed the existence of treeless 
steppes on the Apsheron in the Upper Pleistocene. 

All the authors mentioned above have failed to explain the occurrence in 
one and the same bed of plants, mollusks, insects, birds and small rodents, 
medium-sized carnivores and large, hoofed mammals. Even though the 

138 data confirm the deltaic origin of the bed and the transport by river of 
animal remains, other questions, such as the causes and places of death 
and the distances traversed by the remains, have not been answered. 



132 



139 



'kJ--4 



\У1 



FIGURE 65. Upper part of the 
stem of P h ra gm i t es com- 
munis with closely spaced 
internodes from the bone- 
bearing bed at Binagady 



However, an analysis of the available geological 
and biological data can give a true picture of the 
Binagady site at the time of animal deposition 
and of the landscape of the Apsheron in the 
Pleistocene. 

In addition to Binagady, we studied cases of 
mass death among insects, birds and animals on 
the Apsheron and in Kabristan and the conditions 
of continuous accumulation and burial of their 
remains . 

A study of geological sections in both older and 
newer excavations showed that a lacustrine or 
river origin of bone -bearing bedD is doubtful. 
The osseous strata are always found beneath 
asphalt bedC, and when С wedges out, the organic 
remains disappear. This can only mean that the 
oil-asphalt provided the prerequisite condition for 
the accumulation of organic remains. 

Observations in Kabristan near Khurdalan, 
Kir-Kishlag and Beshbarmak show that 
incrustations and accumulations of asphalt occur 
on mountain slopes and in depressions, near 
peculiar fissures from which the oil is forced to 
the surface. Gas and salt water are ejected, in 
addition to the asphalt, by the mud-oil craters 
or wells. The margins of the latter are overgrown 
by sparse reeds (also found in Binagady, Figure 65), the green of which, 
in the spring, makes a remarkable contrast to the yellow background of 
semidesert, mountains and salinas. The wells attract the cattle, and at 
Beshbarmak cows and asses died after sinking into the thick mud of the well. 

Rainv^ater often accumulates in the asphalt crusts. When the asphalt 
is hot and soft, the water forms lenses which are fatally attractive to 
animals in periods of drought. Asphaltic pools such as these annually trap 
many animals, and remains of various sizes, from that of mouse to 
that of goitered gazelle and ass, can be found in them. The decaying remains 
sink into the asphalt and the bones and mineral particles brought by winds and 
streams become incorporated in the silty beds underlying the asphalt. 

It is not essential that the asphalt pools be located in large reservoirs, 
valleys or guile ys for animal remains to accumulate in the asphalt. On the 
contrary, asphalt located on gentle slopes or even on the crest of a ridge 
is guaranteed long and effective action, because on such terrain it is not 
buried under sediments (Vereshchagin, 1951b). 

These Recent examples indicate that animal death in the Pleistocene 
might have occurred at or near the site of deposit, without necessarily 
involving any transport over long distances. 

Over the centuries, the sticky surfaces and oily freshwater pools created 
by the heavy Binagady oil seeps on the shores of the marine bay probably 
became poisonous traps and burials for a variety of animals (e.g., hares, 
jerboas, gerbils, voles, hamsters) and for many types of birds and insects. 
Heavier animals such as rhinoceroses, horses and bulls would have been 
supported by thicker layers of asphalt. This herbivore carrion undoubtedly 



133 



attracted carnivores and scavenger birds which accounts for the many bones 
of foxes, wolves and hyenas in this area as compared with their paucity 
on the river shoals. The author has described elsewhere the pursuit by foxes 
of birds which were trapped in the asphalt (\'ereshchagin, 1946a). 

Transport of bodies by w^ater with decomposition occurring at the site 
of deposit could have taken place only on a small scale. 

The absence at the site of mesophilous vegetation of forest and meadow 
animals and freshwater mollusks and fish would indicate that the streams 
which carried the bodies must have been small and ephemeral. They were 
probably short-lived torrents originating not more than a few kilometers or 
more likely, a few hundred meters distant from the site. 

Bed (with coquina and juniper trunks) was probably formed during a 
marine transgression over a shallow^ estuary. It covered the bitumen and 
contributed to the consolidation of the bone-bearing bed. During the erosion 
of the southern part of the osseous lense, it would have been possible for 
bone fragments to be redeposited and incorporated into bedB. Tree trunks 
carried into the bay by waves were also preserved in bedB (Figure 64). 

TABLE 16. Age groups of wolves at the Binadagy locality 



Number of specimens 




absolute 


percent 


Age 


1 


0.8 


Approximately 1 month 


20 


16.6 


3.5-4 months 


15 


12.5 


7-8 months 


61 


60.8 


1-3 years 


20 


16.6 


3-5 years, with slightly worn teeth 


3 


2.5 


Over 5 years, with heavily worn teeth 



Such biological data as the seasonal mortality pattern and the composition 
of animal populations can contribute greatly to an understanding of the 
nature of the Binagady locality. Clearly, the insects and pear fruits could 
only have been trapped in the asphalt in the warm seasons of the year. The 
incidence of migratory birds both from the south (spoonbill, purple heron, 
black stork) and from the north (snowy owl) indicates that bird deaths 
occurred during both the summer and winter seasons. 

Death occurred at all ages, from infancy to senility, advanced age 
being indicated by some individuals with molars completely worn down to 
the base. Wolves are a particularly clear example. Because of the greater 
size and strength of the wolf bones, it is easier to extract them from their 
matrices than bones of other species, and thus to form estimates of the 
size of the wolf population. 

The age groups of wolves based on their mandibles are given in Table 16. 

The numbers clearly show that animals of all ages, not only the very 
young and very old, were buried in the Binagady beds. Data in Table 16 
indicate that the main part (67.4%) of the fossil wolves are adults 
(Figure 66). In other words, the dead assemblage represents not a selected 



134 



140 „^ age group, but a cross-section of the 

animal population which lived in the 
area throughout most of the year. 

Analysis of other carnivore fossils 
also indicates that the animals 
perished throughout the warm season 
of the year (Vereshchagin, 1951b). 

Taken together, these facts show 
that the Binagady locality itself 
provided the single cause of death at 
this site, rather than at some 
other locality. 

From these preliminary 
assumptions it is possible to evaluate 
with greater certainty the faunal 
complex at Binagady, as well as the 
ecology of that period in eastern 
Transcaucasia. 

According to calculations by 
Burchak-Abramovich and Dzhafarov 
(1953), nearly 8,213 bones of at 
least 412 individual ungulates were 
collected during the 1938-1942 
excavations. Carnivores are 
represented by 2,043 bones of 11 
species and at least 440 individuals, 
lagomorphs and rodents by 
approximately 2,300 bones of 
15 species and at least 180 
individuals, and insectivores by nearly 
50 bones of three species. 

Bird bones total at least 30,000. 
Hundreds of bones of turtles, lizards 
and snakes, many thousands of insect 
remains (mostly beetles), and 
hundreds of land mollusk shells have 
been found. Thirty-nine species of 
mammals have been identified (Boga- 
chev, 1938b, 1940b; Argircpulo, 1941b; 
Vereshchagin, 1948,1949c, 1951b; 
Burchak-Abramovich, 1951c, 1952d; 
Dzhafarov, 1950, 1955; Alekperova, 
1952, 1955; I. Gromov, 1952, and others). 

We estimated the number of 
common forms from individual 
mandibles and the number of rare species from all the bones. The statistics 
on ungulates were taken from Burchak-Abramovich and Dzhafarov (1953). 
The data in Table 17 is taken from collections which, in 1945, were in the 
142 museums of Baku, Leningrad and Moscow. The quantities shown for bones 
and individuals reflect not only the actual number of animals, but also the 
method of extraction from the rock matrix. For small species, the 
quantities shown are less than the actual numbers. 





*4C^ 



FIGURE 66. Mandibles of wolves of varying age 
groups from Binagady 

(Illustrates Table 16) 



135 



(141) 



TABLE 17. Mammalian species and number of individuals from the Binagady locality 



Species 


Number of 
individuals 


Species 


Number of 
individuals 


Insectivora 


3 
3 
1 


Mesocricetus auratus plani- 
cola 


3 


Crocidura aff. russula 

Hemiechinus aff. auritus 


Meriones erythrourus inter- 
medius 


12 


Erinaceus aff. europaeus 


Allactaga elater dzhafarovi 
A.williamsi 


4 




7 

120 

1 

85 

125 

20 

4 

13 

55 

11 

2 

4 


10 




A.jaculus bogatschevi 

Dyromys nitedula 


26 
1 






] 


Canis lupus apscheronicus . . . 




\ 70 




M apsheronicus 


J 




2 




Ellobius aff. lutescens 

Arvicola cf. terresiris 

Hystrix vinogradovi 




V. vulpes aff. alpherai<yi 


10 
1 


Ursus arctos binagadensis ... 


25 




177 


Meles meles aff. minor 


Perissodactyla 

E q u u s aff. hidruntinus 

E. ca ba llus subsp 








Acinonyx jubatus '. . 


73 




154 


Subtotal 


440 

4 


Rhinoceros binagadensis . . . . 


31 








Subtotal 


258 


Lagomorpha 
Lepus europaeus gureevi .... 


Artiodactyla 
Sus apscheronicus 






4 

2 
3 

8 


11 


Roden tia 
Mus musculus 


Megaceros cf. euryceros .... 
Cervus elaphus binagadensis 
Saiga tatarica binagadensis 
Ovis cf ammon 


2 
52 
82 




1 


Apodemus sylvaticus 




6 


Cricetulus migratorius 






Subtotal 


154 










Total 


1,040 



TABLE 18. Avifauna, by orders, from the Binagady locality 



Order 


Number of 
species 


Order 


Number of 
species 




22 

21 

17 

9 

8 

7 




6 




Steganopodes 


2 


Accipitres . 


Lariformes 


2 




Pterocletes 


1 




Columbus [?] 


1 






1 









136 



Characteristic bones are shown in Figxires 67, 68, 70 and 71. 
The total number of bird species, as given by Serebrovskii (1948) and 
Burchak-Abramovich (1951c), is 97 (Table 18). 

Among the reptiles, bones of the grecian tortoise (T e s t u d о graeca 
L. ) and snnall unidentified lizards are known. 

Insect remains are abundant, but only beetles are well preserved. 
A. Bogachev (1947) has described 107 species of beetles of the following 
families: 

Cicindelidae Dermestidae 

Carabidae Hydrophilidae 

Dytiscidae Coccinellidae 

Cyprinidae Tenebrionidae 

Staphylinidae Cerambycidae 

Silphidae Chrysomelidae 

Histeridae Curculionidae 

Buprestidae Scarabaeidae 

Kirichenko has identified six species of bugs of the Thyreocoridae, 
Pentatomidae and Gerrididae families. 





FIGURE 67. Binagady 

Mandibles: 1 — H em i e ch i n us auritus; 2 — Erinaceus europaeus 

Remains of flies, fly puparia, butterflies, dragonflies and grasshoppers 
are poorly preserved and difficult to identify. Shells of mollusks belong to 
local land species of He lix, Helicella and Pu p i 11 a. 

Plants identified by Petrov (1939) and the author (Vereshchagin, 1949c) 
are represented by the following forms: 



Pirus salicifolia 
Pistacia cf. mutica 
Vitis cf. silvestris 
Z oz im ia sp. 
Tragacanthum sp. 



Juniper us polycarpos 
Punica granatum 
Allium sp. 

Phragmites communis 
Colchicus sp. 
Scirpus sp. 



Bogachev (1940b) has also reported A Ih a gi , Isatis, Artemisia, 
Salsola and Tamarix from the bone-bearingbed. However, these finds 
have not been confirmed. 

Some of the tree remains listed above come from coquina bed B, and 
thus are younger than the main bone -bearing bed. This does not, however. 



137 



143 exclude the possibility of a stable xerophilous landscape in eastern 
Transcaucasia. 

The paleontological material from Binagady is so complete that one can 
speak of entire Pleistocene complexes of different taxonomic groups and 
biocenoses of Transcaucasia. 

An ecological analysis of mammal assemblages reveals specific features 
of the Apsheron landscape, climate and zoogeography. Changes which 
have occurred since the Pleistocene in faunal species composition should 
be projected to the level of orders (Tables 19, 20, 22, 24), taking into 
account the Recent zoogeography of the northeastern and southeastern 
periphery of the Bolshoi Range. 



Insectivora 

The absence of shrews and moles from the Pleistocene of Apsheron 
indicates a xerophilous landscape. At present, the long-tailed white -toothed 
shrew inhabits forests and semideser-t, and the long-eared hedgehogthe steppe 
and semidesert, sometimes frequentingsparse, arid forests. The European 
hedgehog lives in damp forests and in the dry bush of the semidesert. 

TABLE 19. Comparison of Quaternary insectivora species in the piedmont of the eastern Caucasus 





Pleistocene 


Historical time 


Species 


Apsheron 


Apsheron and 
Kabristan 


Dagestan 
piedmont 


Crocidura russula 


+ 
? 
? 

+ 

+ 


+ 
? 

+ 
+ 
+ 


+ 




+ 


Suncus etruscus 


. 


H e m i e ch inus auritus 


+ 




+ 






Total 


3+ 2? 


4+ ? 


4 







Note, The query (?) in this and similar tables indicates species likely to occur within the given period 
of time. 



Carnivora 

The absence of forest species of carnivores (pine marten, European 
wildcat, lynx) excludes the possibility of extensive forests on the Apsheron. 
The lack of otter and mink indicates that there were no rivers or reed- 
grown lakes rich in fish. Since weasels and stone martens are rare on 
the Apsheron in present time, they may not have been preserved. The 
presence of European brown bear (at present found in the sparse, arid 
forests of Transcaucasia) suggests that the ancient Apsheron was a dry 
plateau with juniper and pistachio forests growing along the creek beds. 



138 



144 The presence of small fox and corsac fox is further evidence of the resemblance 
of the ancient landscape to the present. The incidence of cave bear is notable 
and can probably be attributed to the dry climate. 

Fast-moving carnivores were also common: wolf, fox, corsac fox, cave 
hyena, badger, Panthera leo, cheetah, African wildcat. These animals 
are adapted to the steppe or to brush and wooded areas of semidesert. The 
climate of the region might have been colder than it is in the present. 
However, the snow cover must have been thin for the African wildcat, corsac 
fox and fox to have been able to hunt. 

In order to clarify the nature of the deposit and the evolution of species 
assemblages from the Pleistocene to the Recent, the number of carnivores 
at Binagady was compared with estimates of Recent carnivores based on 
contemporary commercial game hunting in the Shemakha area near Baku. 
The study revealed some similarities and some differences in the ecology 
of the two assemblages, and changes in the proportions of five common 
forms: fox, wolf, corsac fox, badger and jackal (Table 21). 

A comparative examination of carnivore species of both periods confirms 
the fact that the characteristically mesophilous Shemakha forest of the 
Recent, which is the habitat of martens and cats, is not at all similar to 
the Binagady trap and mountains. 



Lagomorphs and rodents 

The rodent species composition was almost twice as rich in the 
Pleistocene (Table 22) as in the Recent on the Apsheron. Their ecological 
grouping resembles that of the Recent Dagestan piedmont and indicates 
that, in Pleistocene time, Apsheron was a piedmont steppe marked by small, 
dry forests and some lakes which were overgrown by reeds and cattails. 

Common hamsters live in an environment more mesophilous than the 
one that exists at present on the Apsheron (see Chapter III). The presence 
of the common and the Apsheron vole is another indication that the habitat 
was more mesophilous than now. In contemporary distribution patterns, 
the lower boundary of the common vole lies in the piedmont brush near 
Shemakha at an altitude of 700 m, whereas the pine vole does not descend 
below approximately 1,000 m. 

Other species (hare, migratory hamster, steppe mouse, common 
vole, red-tailed gerbil, jerboa, m.ole vole and porcupine) form an 
assemblage characteristic of the present-day xerothermic valleys of 
northern Iran and Karabakh, which are covered with spiny astragali, 
willow -leaved pear, juniper and hawthorn. 

Comparison of the Binagady rodents trapped in oil pools with those 
caught on the Apsheron in the Recent by eagle owls and by personnel of the 
Azerbaidzhan anti -plague station, indicates that in the Holocene the 
proportion of xerothermic forms gradually increased at the expense of 
mesophilous forms, which ultimately disappeared (Table 23). 

Man is not responsible for the observed decrease of rodents from the 
Pleistocene to the Recent (see Chapter VII). 



139 



145 




FIGURE 68. Binagady 

1, 2 — skull of Vulpes corsac; 3, 4 — skull of V. vulpes aff. alpherakyi; 5, 6 — skull of С a n is 
lupus apscheronicus; 7, 8— canines ofUrsus arctos binagadensis; 9-12 — skull, atlas and 
humerus of С rocu ta spelaea; 13- 17 — skull and humerus of M e les meles aff. minor; 18 — 
skull of Vormela peregusna; 19, 20 — jaws of young A cinony x aff. j u ba lus; 21 — humerus 
of Fel is aff. ly b i ca; 22, 23 — jaws of young and 24 — jaw' of adult Panthera spelaea 



140 



(148)tABLE 20. Changes in the species composition of the Quaternary carnivores in the piedmont of the eastern 
Caucasus 





Pleistocene 


Historical time 


Species 


Apsheron 


Apsheron and 
Kabristan 


Dagestan 
piedmont 


Canis aureus 


+ 
+ 
+ 
+ 

+ 
+ 
+ 

+ 
+ 

+ 
+ 


+ 

+ 

+ 
+ 

+ 
+ 
+ 

+ 

+ 

+ 
+ 
■> 

1 
+ 




Canis sp 




C. lupus 




Vulpes corsac 




V. vulpes 




Hyaena striata 




Crocuta spelaea 




Ursus arctos ' 




Vormela peregusna 




Mustela nivalis 




M, lutreola 




M, eversmanni 




Lutra lutra 




Martes foina 




Meles meles 




Panthera spelaea 




P. leo 


■> 


P. pardus 




Felis lynx 




F.silvestris 




F. lybica 


> 


F. chaus 




Acinonyx jubatus 


5 






Total 


11 


12+ 4? 


17 + 3? 







TABLE 21. Relative proportions of carnivores in Apsheron (asphalt trap) and in Shemakha area (commercial 
game hunting data) 



Apsheron in Pleistocene 


Shemakha area, 1930-1940 


species 


percentage 

of 
individuals 


species 


percentage 

of 
individuals 


Fox 


28,1 

27.5 

19.2 

13.0 

4.5 

2.9 

2.5 

0.9 

0.6 

0.4 

0.2 


Fox 


75.5 


Wolf 


Badger 


16.4 


Corsac fox 

Badger 

Cave hyena 


Jackal 

Wolf 

European wildcat and jungle cat . . 
Pine marten and stone marten .... 
Bear . ... 


3.5 
2.3 
0.8 


Tiger polecat 


0.7 


Lion 


0.2 


Bear 




0.2 


Cheetah 


Weasel 


0.1 


Spotted hyena 


Otter 


0.04 


Wolf (small) 




04 


Striped liyena 


0.001 




Panther 


0.0001 








Number of species 

Number of individuals . . . 


11 
440 


Number of species 

Number of individuals . . . 


15 
12,890 



141 



146 






10 





FIGURE 69. Binagady 

1, 2 — jaws of Lepus europaeus aureevi; 3-6 — femur, metatarsal and tibia of A 11a eta ga 
elater; 7-10 — jaw, femur, metatarsal and tibia of A. williamsi dzhafarovi; 11-15 — skull, 
femur, metatarsal and tibia of A, jaculus bogatschevi; 



142 



(147; 





72 




23 




1 I I -I 








I I 1 ( 




FIGURE 69 (continued) 

16, 17 — maxilla and femur of С ricetulus migratorius argiropuloi; 18-21 — maxilla, 
femur and tibia of Mesocricetus auratus planicola; 22-26 — maxilla, humerus, femur 
and tibia of Meriones erythrourus intermedius; 27-30 — skull, femur and tibia of Ellobius 
aff. lutescens; 31-32 — maxilla and scapula of Hystrix Vinogradov! 



143 



149 JдвLE 22. Changes in the species composition of Quaternary lagomoфhs and rodents in the piedmonts of 
the eastern Caucasus 





Pleistocene 


Historical times 


Species 


Apsheron 


Apsheron and 
Kabristan 


Dagestan 
piedmont 




+ 

+ 
+ 
- 

+ 
+ 

+ 

+ 
+ 

4- 
+ 
+ 
+ 
+ 
+ 


+ 

+ 
+ 

+ 

+ 

+ 
+ 

+ 

+ 


+ 




+ 




+ 




+ 




+ 




+ 


A agrarius 


+ 




+ 


Mesocricetus auratus planicola ... 
M auratus nigriculus* 


+ 




. 


Allactagulus acontion 


+ 


Allactaga elater 


+ 


A williamsi 


_ 




+ 




+ 


Microtus arvalis 


+ 




+ 




. 




. 


E, talpinus 


+ 


Arvicola terrestris 


+ 




. 






Total 


16 


9 


17 







Recent black hamster is a possible descendant of the Binagady species. 



150 Perissodactyla 

The presence of Equus hidruntinus, horse and rhinoceros (Table 24j 
indicates that the Apsheron in Pleistocene time was a grassy steppe with 
bush-grown gulleys. The grasses were probably burnt by midsummer. 



Artiodactyla 

The Pleistocene boar on the Apsheron could have lived in estuarine reeds 
and in the pistachio- juniper forests. Boars now live throughout the year 
in the thin, arid pistachio -juniper forests of Transcaucasia. The presence 
of saiga indicates wide-open areas with steppe vegetation. The Binagady 
Bos and deer might have entered the Apsheron during the spring growth 
of grasses. The decrease of artiodactyls on the Apsheron near the beginning 
of the historical epoch was caused by man. 



144 



(149)^'^'^'"'^ 23. Changes in the species composition and proportions (in fo) of rodents and hares on the 
Apsheron from Pleistocene to Recent 



Species 



Pleistocene, 

caught in 

asphalt pools 



17- 20th 
centuries, 
caught by 
eagle owls 



1939-1940, 

caught in 

traps 



1947, 
caught in 
oil pools 



Steppe vole 

Common vole 

Great jerboa 

Binagady porcupine* .... 
Red-tailed Libyan gerbil. 

William's jerboa 

Transcaucasian mole vole 

Migratory hamster 

Small five- toed jerboa . , 

European hare 

Asia Minor hamster 

Common field mouse . . . , 

Steppe mouse 

Apsheron vole 

Forest dormouse 

Norway rat 



38.6 

14.4 
13.8 
7.1 
5.5 
5.5 
4.4 
2.2 
2.2 
1.6 
1.6 
1.1 
1.1 
0.5 



23.8 



18.1 
9.5 



42.4 
0.2 



5.7 



4.6 

Absent 

Absent 

Extinct 

85.4 

2.0 
Absent 

2.9 

0.3 

0.6 
Absent 

0.2 

3.2 
Absent 

0.1 

0.1 



25.0 



29.2 
20.8 

12.5 
0.3 
4.1 



Number of species . . . 
Number of individuals 



15 
181 



7 
347 



9 
2,473 



6 
24 



Recent Indian porcupine is a possible descendant of the Binagady species. 



TABLE 24. Changes in composition of Quaternary Perissodactyla in 

150) 


he piedmonts of the 


eastern Caucasus 




Pleistocene 


Historical time 


Species 


Apsheron 


Apsheron and 
Kabristan 


Dagestan 
piedmont 


Equus hidruntinus 


+ 

+ 
+ 


+ 
-> 




E. hem ion us ' 




E. caballusgmelini 




E. caballus subsp 




Rhinoceros binagadensis 


- 


Total 


3 


1+ 1? 


2 



152 In studying the Pleistocene landscapes of the Apsheron, birds contribute 
little to our knowledge because of their seasonal migrations. Beetles, 
however, are more reliable. The proportions of beetles found at Binagady 
in aquatic, amphibious and terrestrial habitats respectively is given in 
Table 26 (identification by A. V. Bogachev; author's figures). 

Aquatic beetles, as such, are not indicative of the landscape, because 
of their capability of flying great distances; water-scavenger beetles and 
water beetles have even been known to cross great expanses of desert to 
reach water. According to Bogachev, land beetles in the Binagady 
bone -bearing bed comprise 8% (out of a total of 75.7%) of relatively 
mesophilous forms which are characteristic of the present piedmont brush 
near Shemakha and are absent from contemporary Apsheron. In other 



145 



151 







FIGURE 70. Binagady 

1-4 —M , M , heel bone and hoof ofEquus caballus subsp. ; 5-9 — M^, M^ Ртз-М^, heel 
bone and pelvis ofEquus hidruntinus; 10, 11— skull of Rhinoceros binagadensis 
(top and bottom views) 



146 



(150) TABLE 25. Changes in composition of Quaternary Artiodactyla in the piedmonts of the eastern Caucasus 



153 





Pleistocene 


Historical time 


Species 


Apsheron 


Apsheron and 
Kabristan 


Dagestan 
piedmont 




+ 

+ 
+ 
+ 

+ 

+ 
+ 


+ 
+ 

+ 

? 






+ 


Cervus elaphus 


+ 








+ 


Gazella subgutturosa 


+ 






Bos mastan-zadei 






? 


Bison bonasus 


+ 






Total 


7 


3+ i? 


5+ 1? 







It is possible that the Apsheron boar is ancestral to the contemporary species, 
™' The primitive bull inhabited Apsheron later than the period of the Binagady trap. 

words, the shift in the range areas of beetles confirms a progressive 
development of xerophilous landscapes in the Holocene. 

Thus, both qualitative and quantitative data on the mammal and beetle 
fauna and on the flora show that the climate of the Pleistocene Apsheron 
was colder and moister than in the present. The ecological conditions 
were similar to those which now exist in the thinly forested, arid zone 
on the southern slopes of the Armenian Highland at elevations of 1,500- 
1,600 m, and in the Dagestan piedmont at elevations of 0-500 m. 

The universality of the Binagady assemblage leads to other conclusions 
regarding the stratigraphy, paleoecology and history of the formation of 
the fauna. 

The Binagady faunal complex contains a representative group of species 
widely distributed in the Pleistocene over steppes and forest-steppes, and 
known from Paleolithic and other localities of Europe. These include: 



Erinaceus europaeus 
Canis lupus 
Vu Ipes corsac 
V. vu Ip es 
Crocuta spelaea 
Ursus arctos 
Meles meles 
Panthera spelaea 
Allactaga jaculus 



Apodemus sylvaticus 
Mus musculus 
Microtus arvalis 
Arvicola terrestris 
Equus caballus 
Megaceros euryceros 
Cervus elaphus 
Saiga tatarica 



The origin of most of these species is not yet clear. Only the corsac fox, 
great jerboa and saiga migrated from the north into eastern Transcaucasia. 

From the ecological point of view, the above assemblage is heterogeneous. 
It includes some species which, at present, are adapted to open habitats, 
and some species which live in the forests and moist river valleys of the 
same regions. There are no true xerophilous forms in the assemblage. 

Some of the listed species (fox, wolf, bear, badger, great jerboa) are 
represented at Binagady by southern forms, an index of the antiquity of 
the conditions under which the Binagady faunal complex evolved. 



147 



(152) 




FIGURE 71. Binagady 

1,2— epistropheus and metatarsal of S u s a p s с h e ro n i с us; 3, 4 — metatarsal and humerus 

Ovis aff. ammon: 5-8 — horn, atlas, scapula and metatarsal of Saiga tatarica binagadensis; 

9, 10 — skull of Bos mast a n- zad ei , from above and side (1/20 natural size) (from Bogachev, 1940) 

TABLE 26. Ecological groups of beetle species and numbers of individuals from the Binagady asphalt 



Ecological groups of beetles 


Number 


of species 


Number of 


individuals 


absolute 


percent 


absolute 


percent 


Water 


15 
11 
81 


14.1 
10.2 
75.7 


760 

74 

2,758 


21.1 


Land and water 


2.0 


Land 


76.9 






Total 


1U7 


100.0 


3,592 


100.0 







The species which became extinct are: cave hyena, cave lion, 
horse, giant deer. The ranges of corsac fox, great jerboa and saiga 
have shifted considerably, but only localized changes took place 
in the distribution areas of the common and water voles. 
154 There is a prominent group of Mediterranean species adapted to 

xeromorphic landscapes, such as the upland steppes of southwest Asia. 
These species are: 



148 



Crocidura russula Microtus socialis 

Hemiechinus auritus Ellobius lutescens 

Vormela peregusna Meriones erythrourus 

Felis lybica Hystrix vinogradovi 

Acinonyx jubatus Rhinoceros binagadensis 

Lepus europaeus Equus hidruntinus 

Allactaga williamsi Sus apscberonicus 

Cricetulus migratorius Ovis am mo и 

Mesocricetus auratus ' Bos mastan-zadei 

The group includes species adapted to mixed biotopes (white-toothed 
shrew, long-eared hedgehog, tiger polecat, jungle cat, cheetah, 
William's jerboa, steppe vole and boar), to upland steppes (Radde's 
hamster, migratory hamster and Transcaucasian mole vole), and to hot 
valleys and lowlands (red-tailed gerbil and Binagady porcupine). 

Most of these Mediterranean forms were represented in the Pleistocene 
by distinct subspecies, whose main areas of distribution at that time lay 
definitely to the south of the Caucasian Isthmus. The extinct Pleistocene 
ass (Equus hidruntinus), Binagady rhinoceros and Binagady goat were 
also of southern and southeastern origins. The extinct Apsheron vole, 
which has been identified from its m^olar classification as belonging to the 
European type of subgenus Pit у my s , can be included among the endemic 
species. The range of the Dagestan hamster shifted into inner Dagestan, 
and that of the mole vole, to the mountain-steppe of Talysh. 

Thus, the general evolutionary trend of the faunal complex on the eastern 
Transcaucasian plains was toward either extinction or shifts in distribution 
patterns of relatively mesophilous forms. The disappearance of these forms 
from the area was caused by the development of a drier climate and 
consequently of a more xerophilous landscape. 

The general paleozoogeographical relationships within the Binagady 
complex can be understood when the complex is compared with the 
Pleistocene faunas of the southern part of the Russian Plain and of Southwest 
Asia. 

Some common species of the Middle and Upper Pleistocene of Eastern 
Europe do not occur at Binagady (e.g., mammoth, woolly rhinoceros, 
Bison priscus, cave and other bears). Northern species were also 
absent from the Binagady complex: arctic fox, blue hare, lemming, 
reindeer, musk-ox. 

The absence of these forms may be accounted for in part by the existence 
of a xeromorphic landscape in the eastern Caucasus, and in part by the 
geological age of the locality. 

Tundra species probably did not extend as far south as the area of the 
Binagady trap, at the time it existed. Neither the harvest mouse, striped 
field mouse nor Siberian polecat reached the area from the north. And so 
it was that only some of the faunal elements of the southern Russian Plain 
and of the Iranian-Asia Minor uplands met in Apsheron and in Kabristan 
155 during the Pleistocene. Certain southwest Asian species, listed below, which 
have been found in the Middle and Upper Pleistocene of Palestine, Syria 
and Lebanon, were also absent from the area under consideration: 



149 



Herpestes ichneumon L. Dama mesopotamica L. 

Felis chaus Guld. Gazella arabica L. 

Panthera pardus Schreb. Gazella cf. subgutturosa Guld. 

RattusraitusL. Capraprimigenia Fraas 

Spalax sp. C. beden Schreb. 

Phacochoerus garroda Bate Bubal us sp. 

Hippopotamus amphibius L. Procavia cf. syriaca Schreb. 

The distance between Apsheron and Syria and Palestine indicates that 
the differences in composition of the fauna have zoogeographical significance 
which persisted through geologic time. 

A comparison of the Binagady complex with the Recent fauna of the eastern 
Transcaucasian plains shov.'s the absence in the earlier period of a number 
of eurytopic species — species which probably penetrated the Apsheron at a 
later time, some still extant, such as Suncus etruscus, Canis 
aureus, Hyaena hyaena, M u s t e 1 a nivalis, Felis chaus, 
F. lynx, Panthera pardus, Rattus norvegicus, Gazella 
subgutturosa and Martes foina, and some recently extinct, such as 
Panthera leo and Equus hemionus. 

Thus an interesting feature becomes apparent: a number of southern 
thermophilous species migrated to eastern Transcaucasia, while a number 
of European steppe species retreated into eastern Ciscaucasia in the period 
following the time of the Binagady trap. 

All of which emphasizes the specific features of the Binagady complex 
and the relatively short period of time during which the asphalt trap was 
operative in the Pleistocene. 



Stratigraphic correlations and age of the Binagady complex 

The age of the Binagady locality is significant in tracing the origin of 
Quaternary fauna and the morphogenesis of mammals. 

Atteriipts at dating the bone-bearing bed and the faunal complex of 
Binagady have been based on the estimated age of the old Caspian terrace 
and of the orogenic compression of beds which was accompanied by an oil 
flow, and on paleozoogeographic correlations. 

Assuming the age of the terrace underlying the bituminous beds to be 
Khazar, Bogachev (1939, pp. 47-51) has dated the Binagady locality Riss- 
Wiirm. He has also related the compression of the productive beds and the 
oil flow to the so-called "Kalinovka" phase of folding. According to Reingard 
(1937) this phase occurred at the beginning of the second Riss glaciation. 
According to Vardanyants (1948, table 3, pp.104, 125), considerable 
compression occured in the pre -Khazar and pre-Khvalynsk orogenic phases, 
and later in the Neo-Pleistocene and Holocene, resulting in the formation 
of folds on the Apsheron and the development of mud volcanos. According 
to this scheme, the Binagady bed might have formed either in pre -Khazar 
or pre-Khvalynsk time. 

Bogachev places the tir.ie of truncation of the Binagady fold at the Khazar 
transgression, and concludes from this that the oil which appeared during 
156 the Kalinovka orogenic phase could not have been instrumental in trapping 
and preserving plants and animals, because the oil would have disappeared 



150 



prior to the Khazar transgression. Accordingly. Bogachev (1940b, p. 14) 
postulates the dislocation of Didacna surachanica beds and the 
existence of two Kalinovka compressive phases. 

An assumption that the Binagady fold was truncated by the Baku sea 
would, in fact, lead to a correlation between the formation of the asphalt 
trap and the pre-Riss and the Kalinovka orogenic phases, and would affect 
an estimate of the age of the locality. 

The peculiar features of the Binagady complex, as compared with the 
Quaternary complexes of the Russian Plain, complicate biostratigraphic 
correlations. Moreover, the difficulties are compounded by the lack of 
index fossils, such as elephant and bison, at Binagady. 

Gromov (1948, p. 424) has dated the Binagady complex Upper pre -Glacial, 
correlating it with the Middle Quaternary "Khazar fauna of the Volga 
region," which Nikolaev (1937) dated Mindel-Riss. 

The Burchak-Abramovich—Dzhafarov summary (1955) indicates that a 
majority of paleontologists accepted the Bogachev estimates without verifying 
them independently. 

In a study of carnivores from the Binagady asphalt (Vereshchagin, 1951b), 
the author employed the same age estimate with a disclaimer that the dating 
can be improved through more complete excavations and further study of 
the relationships of the continental and marine beds at the Binagady mound. 

A comparison of the Binagady species with lists of species from the 
Paleolithic caves of Palestine, Syria and Lebanon (Bate, 1937; Picard, 
1937) shows that the greatest similarity between the faunas of the two regions 
existed during the Mousterian and Aurignacian, i.e., in the Upper 
Pleistocene. 

We determined the residual organic content of the Binagady horse and 
saiga bones by calcination (as used by Pidoplichko, 1952). The results 
showed that upon oil extraction by hot alkali and benzine, the calcination 
index was in the 480-520 range, which is close to the index of the bones 
of the Khazar complex. 

V. V, Cherdyntsev's attempt at a radioisotope determination of the age 
of the bones was unsuccessful. 

On the basis of all the paleobiological and geological data, the Binagady 
complex must have existed and been buried at the beginning of the Khazar 
transgression, i.e., the end of the Middle Pleistocene. In Reingard's 
scheme, this corresponds to the lower part of the Middle Caspian beds and 
the second Riss glaciation ("Rn"); in Gromov's, it corresponds to the 
Riss, "Qj\". 

How long the Binagady complex existed within the framework of the faunal 
composition described is not accurately known. Some species of the complex 
continued to live even in the Khvalynsk age, while others were replaced. 
An example of a later deposit where the Bos of the Asian type (Bos 
mastan-zadei) shows replacement by the European tur, while the saiga 
continued to exist, is another asphaltic locality, Kir-Maku. 
157 Morphological development of eastern Transcaucasian animals during 
the Upper Quaternary is discussed in Chapter IV. 

The results of the taphonomical and zoogeographical analysis of the 
Binagady burial may be summarized as follows. 

1. The Binagady mammals lived in a dry steppe and in sparse, arid 
forests which grew in guUeys and in areas of placer rock. The climate of 
the region was somewhat cooler and more humid than in the present. 



151 



2. Animals of various ecological types perished in the liquid oil and 
viscous asphalt, as well as in mudflows. 

3. The animals were buried in the asphalt sand and fine-grained coquina 
at or near the site of death. The asphalt beds were later covered by eolian 
sandy coquina and lacustrine -estuarine sediments. 

4. On the basis of the occurrence of Khazar molluscan index-species 

in the Early Caspian terrace, the Binagady bone -bearing bed on the terrace 
margin can be dated post-Baku age. 

By relating the profuse oil flow and the deposit of animal remains with 
the orogenic phases, the Binagady faunal complex can be dated as either 
Lower Khazar or Lower Khvalynsk. 

5. A comparison of the Binagady faunal complex with the Quaternary 
complexes of the xerothermal zone of southwest Asia (the Paleolithic of 
Syria, Lebanon, Palestine) indicates that the Binagady complex correlates 
with the Mousterian culture. The coefficients of the residual organic matter 
of the Binagady bones are similar to those of the Khazar bones of the Volga 
region. 

6. Of the 39 mammal species in the Binagady complex, 19 (48.7%) are 
Pleistocene species found widely distributed on the plains of Western and 
Eastern Europe. Another 19 (48.7%) are Pleistocene species of southwest 
Asian and Iranian origin. Only the one rennaining species (2.6%) can 
possibly be endemic. 

The faunal complex probably did not include arctic or mountain aninnals, 
or cave bears, elephants and bison. During the time of the complex, the 
ranges of several animal species of the steppes of the Russian Plain and 
of the upland steppes of southwest Asia met on the Apsheron. Fossiliferous 
localities of younger age on the Apsheron contain faunal assemblages which 
are not as universal as that of Binagady. 



Kir -Maku 

The locality is ll km south- southeast of the village of Dygya, in 
the center of the Apsheron Peninsula, on the slope of the Kir-Maku asphalt 
mound. 

The Kir-Maku site was discovered in 1870 by Shtukenberg, who 
collected and gave to the Kazan University a skull of a fenaale saiga and a 
wolf mandible in an asphalt matrix. 

According to the map of Volarovich and Lednev (1913-1929), the site 
lies 100 m above sea level; the mound is nearly 1.5 km wide at the base 
and is 60-70 m high. The top of the hill is composed of Early Caspian 
deposits, underlain by almost vertical middle productive sands (Middle 
158 Pliocene). Two buried or half-buried asphalt flows lie on the northern slope 
of the hill; one asphalt flow occurs on the southeastern slope. A bone- 
bearing kir layer occurs somewhat below the top of the hill, at a depth of 
2-5 m and covers an area of 1.5-2 hectares. A generalized stratigraphic 
section through the bituminous beds of the Quaternary at Kir-Maku as given 
by Sultanov (1947) is shown in Figure 72. 



152 



2Р-5Л 



The bitumen in this area often occurs in vertical veins, nearly 20 cm 
thick at the base and 1.5-2 m long. However, the main mass of the bitumen 
lies on a continuous layer, probably of flow origin. Bones occur in both 
the stratum and the veins. The oil which was forced to the surface by 
compression of the beds probably formed sticky crusts and pools which were 
later covered by sand and dust. Only fragmentary skeletons and skulls are 
found, as a rule, in the bitumen, probably remains of animals that died 
trapped in the asphalt crusts. 

The Kir-Maku locality has something in common with the 
travertines of Mount Mashuk in the northern Caucasus. However, the 
asphalt crusts probably acted as a more efficient trap. 

Bogachev (1924, 1925b) reported a skull 
of Bos primigenius that was 
indistinguishable from a species which lived 
on the Russian Plain during the Pleistocene 
and Holocene. He dated the bituminous salina 
near the village of Dygya post-Baku and 
thought that the Bos died and was preserved 
during a period of increasing humidity and 
expansion of the steppe flora on the Apsheron 
Peninsula in Khvalynsk time. 

In 1940-1941 a large number of bones of 
Bos primigenius, wolf, fox, cave 
hyena and vulture were found in the process 
of bitumen-mining. The miners buried the 
bones in a specially prepared trench 
(Bogachev, 1944; Sultanov, 1947). 

The Kir-Maku beds are younger than the 
Binagady beds. The age of the relief at 
Kir-Maku can be readily determined, since 
the bituminous bed is concordant with the 
slope . 

The Bos primigenius from Kir- 
Maku is a younger type which replaced the 
Binagady form in the Upper Pleistocene of 
eastern Transcaucasia. The Kir-Maku 
Pleistocene assemblage still included saiga 
and possibly hyena of the Binagady type. 

Fossils found in the bitumens of the 
northeastern part of Artem Island belong to 
159 f^:f:i^:f:^:^:f::f:f''f/A the Upper Pleistocene. These include two 

red deer skulls with broken antlers which 
were placed in the Museum of Natural 
History in Baku (Bogachev, 1938a). Other 
bones of red deer, and several bones of horse, 
wolf and large birds vv^ere collected at the 
same locality in 1944 from old caches 
(Burchak-Abramovich, 1951c). 

The age of the bones from the island 
approaches those of Kir-Maku, as nearly 
as can be inferred from the state of preservation of the material. 




FIGURE 72. Section through the asphalt- 
bearing beds at Kir-Maku 

a — surface loams; b — sandy- clayey 
sequence; с — "rich" asphalt bed; d — 
productive sands. On the right — 
thickness in meters 



153 



HISTORICAL EPOCH 

Broadleaf forests 



Saline semidesert 
of uplands 




10 20 30 io 50 iO 70 80 SO 100 1Ю 120 130 lU0 150 160 



km 



MIDDLE PLEISTOCENE 



Piedmont 



Binagady locality 



Arid thin 




10 20 30 M) SO 60 70 80 30 100 110 120 130 m 150 160 Vru 



FIGURE 73. Changes and migration of landscape zones since the Middle Pleistocene on the Apsheron 
Peninsula, based on the study of the Binagady locality 



160 



Deer and horses probably lived on Artem Island during the relatively 
humid, cold period in the Top Pleistocene. At that time ungulates migrated 
freely over the Apsheron and to nearby islands, particularly during the 
spring growth of ephemeretum. Occurrences of animals on the islands, 
therefore, do not indicate a regression of the Caspian, since the animals 
could swim across the narrow straits. 

The Pleistocene development of landscapes and faunal complexes on the 
eastern Transcaucasian plain can be described from data at hand, as 
follows. 

The development of the xeromorphic landscapes which predominated 
in the Lower Pleistocene probably began in the Middle Pliocene (the time 
of the productive beds). During the Baku transgression the shores of Kura 
bay wei^e inhabited by rhinocei-oses, E 1 a s m о t he r i u m , horses, camels, 
and, of the carnivores, hyenas which closely resembled the spotted or 
African brown hyena. 

A distinct complex of mammals developed on the plains of eastern 
Transcaucasia toward the Middle Pleistocene, consisting of species widely 
distributed on the steppes of the temperate zone of Europe and on the 
upland steppes of southwest Asia. 



154 



By analogy with western Transcaucasia, it can be assumed that the 
mountains of the Bolshoi Range were inhabited by European brown bears 
and cave bears, martens, lynxandherds of mountain goats and chamois. 

Glaciation over the eastern spurs of the Bolshoi Range, while probably 
not extensive, was sufficient to create a number of changes in the fauna 
of the plains. The general landscape was characterized by sparse 
forests and steppes. 

The Bos of the Asian-Indian type was replaced by the European tur. 
Argali and giant deer probably disappeared. A number of Pleistocene 
species of the steppes and forest-steppes of Eastern Europe penetrated 
Transcaucasia, and possibly the Iranian Plateau, during a cool, humid 
climatic phase in the Upper Pleistocene. During this stage the altitudinal 
vegetation zones probably migrated down the mountain slopes (Figure 73). 
This, however, did not interrupt the continuity of the xerophilous fauna 
of southwest Asian origin. 

The East European steppe animals (corsac fox, great jerboa, saiga) 
retreated to the north only at the very end of the Pleistocene during a new 
xeric climatic stage. 

A new migration of southern species into Transcaucasia occurred at 
the beginning of the xeric Holocene epoch. Some of the species migrated 
as far north as the Ciscaucasian plains. 



Lesser Caucasus upland (southern 
Transcaucasia) 

The high plateaus of southern Georgia, Armenia and Karabakh were 
taphonomically in a special position in the Pleistocene. Frequent mudflows, 
landslides, rockfalls and volcanic eruptions, associated with the 
mountainous terrain, caused death among animals. However, their incidence 
as fossils is rare. The known Pleistocene localities are situated on the 
margins of broad river valleys and in sink holes of high plateau lakes. The 
bones were deposited in river and mudflow fanglomerates, in diluvial 
sediments near mountain slopes and in lacustrine sediments. 

Although Paleolithic material is plentiful on the Armenian Highland, 
the instrumentality of man in the accumulation of animal bones was 
negligible (Zamyatnin, 1947; Panichkina, 1948, 1950a, 1950b; Sardaryan, 
1954), probably because of the erosion of most of the Middle Paleolithic 
caves and open sites. 

A site containing bones of Lower Pleistocene mammals in ancient 
lacustrine deposits near Akhalkalaki is particularly interesting. The 
locality was discovered by Vekua in 1958 and from it he has identified the 
following species: Canis sp. , Crocuta spelaea, Meles cf. meles. 
Marmot a sp. (small form), Elephas cf. trogontherii, Equus 
sp. , Rhinoceros etruscus. Hippopotamus sp. (large form), 
Megaceros sp. , Bos cf. primigenius. The presence of the 
hippopotamus and of an unusual marmot in the assemblage emphasizes its 
southern aspects. The incidence of both forms indicates that the landscape 
161 of the area at the dawn of the Anthropogene was comprised of low plateaus 
with a well-developed river network and warm climate. * 

' The age of this assemblage may be older, possibly Upper Pliocene. 

155 



FIGURE 74. Skull of Bos primigenius from Bayandur on Armenian Highland 

A large accumulation of bones was discovered in 1928 in the sand-gravel 
quarries near the Kazachii post on the southern outskirts of Leninakan, of 
which up to 300 kg were removed to the Yerevan Museum (Bogachev, 1938c, 
1938d; Avakyan, 1948). Bogachev has identified the following species in 
the material collected: Bison priscus, Megaceros sp. (hiberni- 
cus?), Cervus elephas maral, Equus caballus. Bones of an 
elephant (Elephas cf. armeniacus) were taken from a lower bed 
which has been assigned to the Upper Pliocene (Apsheron age or equivalent). 
The rest of the bones were dated Pleistocene. Avakyan and Burchak- 
Abramovich (1945) have identified the camel Camelus knoblochi in 
the material and the following species occurring with it: Equus 
stenonis, Rhinoceros cf. mercki, Elephas m e r i d i о n a 1 i s , 
E. trogontherii (?), Cervus sp. , Bos sp. 

Elephant teeth from near Leninakan, described by Sardaryan (1954), 
indicate that the predominant species in the collection is Elephas 
trogontherii, which closely resembles the Tiraspol E. wusti and, 
to a lesser extent, the elephants from the Taman Peninsula conglomerates. 
A fragment of skull of Bos primigenius from this locality (collections 
of the Museum of Georgia) has been described by Burchak-Abramovich 
(I951d). 

Another fragment of Bos primigenius skull (Figure 74), taken from 
near the village of Bayandur in the Leninakan area, has been described 
by Gromova (I93l). Even this brief review of the material clearly shows that 
fossils in the alluvial deposit near Leninakan are of varying ages. The 
deposit strongly resembles the gravels of the inclined Ciscaucasian plains. 

A Lower Quaternary faunal complex, with such indicative forms as 
Elephas trogontherii, Equus stenonis and Rhinoceros 
mercki, is distinguishable in the material collected and shows a correlation 
of the Leninakan freshwater deposit with the upper part of the Psekups 
gravels and the lower beds of the Girei quarry on the middle Kuban. 

The Lower and Middle Quaternary bones from the sands and gravels 
near the village of Eilas, 12 km southwest of Yerevan, are close in age to 
the material described. In the Adzhi-Eilas site the gray, fine-grained river 
sands containing mammalian bones underlie the 3— 3.5-m-thick gravels and 
loessial loams. The bone-bearing lenses are up to 1-1.5 m thick. Some of 
the bones had been redeposited, being apparently derived from older 
lacustrine sediments. The beds were dated as Riss age by Avakyan (1946), 
The fragments have lost most of their organic matter. The surrounding 
matrix of the horn fragments is "chalky" and "marly. " The enamel of the 
deer teeth is grayish brown from staining of the dentine. The material which 



1704 



156 



jgnwe studied includes teeth of E le p h a s t r о go nt he t i i , a fragment of 
the skull of "diluvial-type" Bos, Bos trochoceros, fragments of deer 
horns and teeth, Cervus ex. gr. elaphus and Dama cf. mesopota- 
m i с a (Figure 75). 




FIGURE 75. Adzhi-Eilas 

1 — skullof Bos trochoceros; 2, 3 — fragment of horn and jaw of D a m a cf. 
mesopotamica 



Younger, Upper Pleistocene bone deposits, partly created by primitive 
hunters, are so far known only in the northern sections of the Lesser 
Caucasus upland. There is a record of rodent bones, a fragment of a deer 
horn, fragments of tusks and skeletal bones of E le ph as primigenius 
and a tooth ofEquus caballus taken from a depth of 10 m at a site 
on the left bank of the Chichkhana rivulet near Nalband, 30 km north of 
Leninakan (Karakash, 1898). 

Zurtaketi site. This Paleolithic site discovered by Maruashvili 
(1946) on the Tsalka Plateau in southern Georgia was ascribed particular 
stratigraphic significance. The site is located at an elevation of nearly 
1,320 m on the right slope of the Karabulakh River valley (the right 
tributary of the Khram River), 15 km north of Bashkicheti. Fractured bones 
and tools of flint, obsidian and bone reflect human activity at the site. The 
obsidian tools are ten times as abundant as the flint implements. The 
bone and stone tools must have been attached to handles, because they 



157 



resemble darts and arrowheads. From the stone material the site has been 
dated as Solutrean. 

Because he assumed that the primitive tribe was driven off the site by 
a lava flow which burned and covered the cultural layer, Maruashvili thought 
that the uplands relief and the volcanic activity dated from a period later 
163 than that of the site. His age estimates were disproved by Paffengol'ts 
(l95l) who established that the site is located in a collapsed recess of a 
doleritic lava wall. 

The mammalian material collected by Maruashvili in 1945 and 1948 
consisted of molars, incisors and metatarsal epiphysis of a horse, molars 
of an ass, vertebrae of a proto-bison and a fragment of a ram radius. 

Most of the bones are weathered and have slightly dendritic surfaces. 
Their state of preservation indicates a very late stage of the Pleistocene. 

The material from the Zurtateki site, as identified by Burchak- 
Abramovich (1951a), is given in Table 27. In addition, partridge bones 
which resemble rock partridge have also been identified in the locality 
material. 

The species composition of the game of the Zurtaketi people reflects a 
Khram Basin landscape of rolling plateaus covered with meadow -steppe, 

TABLE 27. Mammalian species and number of bones from Zurtaketi site 



Species 


Number of bones 


Number of individuals 


Equus caballus subsp 

E. (A s i n u s ) sp 


14 
3 
6 
1 


3 

1 


Bison priscus 


1 


Ovis cf.gmelini 


1 






Total 


24 


6 







and underdeveloped forest -steppe vegetation in the Upper Pleistocene. 

The asses (kulans) .and small rams were quite common until recently 
in the woodless territories farther south. 

Prehistoric hunters were also active along the upper Khram River, 
where a Mesolithic cave site, approximately 8 m above the riverbed, was 
discovered by Kuftin (1941) near the village of Barmaksyz. 

At this site, a cultural layer of no more than 40 cm in depth yielded 
9,900 broken obsidian tools and 600 flint tools in a 70 m area of eroded 
slope with surface soil up to 80 cm thick. 

The bone material consists of fragments of diaphyses and ungulate teeth, 
stained with humus. 

According to Burchak-Abramovich (1951b), the collections include six 
molars of a small horse and a lower molar of a Bos. 

During the artificial lowering of Lake Sevan, Dal' (1950b) collected 
material along the shores which is the samie geologic age as the collections 
mentioned above. 

The extensive collections are mostly fragments of bones of domestic 
animals of post -Paleolithic age which are discussed in more detail below. 

As far as the state of preservation shows, the oldest bones belong to a 
Bos cf. minutus and a dromedary (Camelus d r о m e d a r i u s), 
164 tentatively identified by Khavescn (l954b) as a wild form. 



158 



The fauna of the Pleistocene beds of southern Transcaucasia is relatively 
poor in species. Nevertheless, there is a regularity in the replacement of 
the faunal complexes one by another, similar to the process which operated 
in Ciscaucasia. 

It is interesting to note that a number of index species characteristic of 
the "cold steppes" of Europe occur in the Pleistocene deposits of the Lesser 
Caucasus plateau. The following species are known: Elephas trogon- 
therii, i^hinoceros mercki, horse, Camelus knoblochi, 
Megaceros, Cervus elaphus. Bos prim, igenius. Of the species 
peculiar to southwest Asia, only fallow deer and mouflon-like sheep are 
known. The southern east Mediterranean aspects of this faunal complex 
were probably better developed among the smaller forms in the Pleistocene. 

The species under study have not yet been classified stratigraphically, 
nor have morphological characteristics been established for the index forms 
which take into account their environmental adaptations. 

As opposed to the Greater Caucasus, the Upper Pleistocene faunal 
complex of the highland region under discussion has not yet been recorded. 
This is partly accounted for by the poor development of the karst, and by 
the fact that the karst itself has been inadequately studied. In the face of the 
scarcity of Paleolithic naaterial, it is impossible to form a judgment of the 
effects of Pleistocene cooling and giaciation of the plateaus on the 
developnaent of the fauna. There is no doubt, however, that the increase in 
water and in mesophilous vegetation on the plateau did not preclude the 
existence of camels in the hot intermontane valleys. The vertical shifts of 
vegetation zones and faunal complexes which occurred during the period of 
cooling in the Pleistocene were probably less extensive than in the Greater 
Caucasus. 

The period of transition to the Holocene on the Armenian Highland is not 
very well known; it is probable, however, that it was similar to that in 
eastern Transcaucasia. 

The stratigraphic and geographic distribution of Pleistocene mammals 
in the Caucasus is given in Figure 2 (see Introduction). 



ECOLOGY AND LANDSCAPES OF THE CAUCASIAN 
ISTHMUS ACCORDING TO PALEONTOLOGIC AL DATA 

A review of the geological data and of Pleistocene fossiliferous localities 
by regions provides a basis for a summary of the specific features of the 
development of Caucasian fauna. 

From the excavations at Kudaro I, the faunal complex which existed in the 
mountains that were uplifted in the Lower Pleistocene can be recognized 
as the direct ancestors of Recent Caucasian species of insectivores, rodents 
and ungulates which are characteristic of the Mediterranean region of the 
Alpine folded belt. From the zoological point of view, the strong 
peneplanation of the Caucasus in the Lower Pleistocene which is assumed 
by some geologists (see, for example, Vardanyants, 1948) is an 
impossibility. The evolution of the Caucasian mountain fauna with such 
165 highly specialized forms as Prometheomys, Microtus nivalis, 
Caucasian goat and chamois took place at the same time as the evolution of 
similar forms in the Alps, the Carpathians and Asia Minor. It is difficult 



159 



to support a case for the origin and development of these mountain species 
only in the Pleistocene, since, by the end of the Middle Pleistocene, they 
had already assumed characteristics which are identified with the Recent. 

The main sections of the Greater Caucasus which had been uplifted during 
the Miocene probably retained their elevation into the Pleistocene, even 
though they might have been somewhat eroded. 

The better-known death assemblages of Middle and Upper Pleistocene 
mammals in the Caucasus reflect a continuing cooling of climate and 
modernization of fauna. 

Elasmotherium and Cervus pliotarandoides disappeared 
in Ciscaucasia and Transcaucasia toward the Middle Pleistocene. Steppes 
and forest-steppes replaced the semideserts and savannahs on the piedmont 
plains and the evolution of landforms promoted further isolation of faunal 
complexes in different landscape zones. At that time, the forest -steppes 
of Ciscaucasia were inhabited by Elephas trogontherii. Rhino- 
ceros mercki, red deer and long-horned bison. In eastern and southern 
Transcaucasia the bison were replaced by В о s primi genius. 

A widespread invasion of eastern Ciscaucasia by some xerophilous 
species of the Aral-Caspian fauna (e.g., little suslik. Central Asian gerbil 
and jerboa) might have occurred very early, perhaps even at the close of 
the Apsheron. It is possible that the ancestors of these species survived 
the Caspian transgressions on lands bordering the east Ciscaucasian bay, 
or that, having disappeared from the area, they returned in new migrations 
during the regressions of the sea. 

A mountain-forest faunal complex, consisting of bear, boar, deer, 
Caucasian goat and bison, existed in western Transcaucasia on the steep, 
southern, seaward slope of the Bolshoi Range. 

The faunal migrants from the Russian Plain were represented by the 
common hamster which appeared as far south as Tsebel'da. 

The low plateaus of Imeretia formed the northwestern boundary of the 
distribution zone of Upper Pleistocene xerophilous species of the southwest 
Asian uplands: Radde's hamster, porcupine, ass, argali-like sheep. Apes 
also survived in the area from the time of the Pliocene. The presence of 
horses and asses indicates a steppe development along the middle section 
of the Rion. 

Similar steppes, covered in places by very extensive lava flows, also 
developed in the Lesser Caucasus uplands and in central Transcaucasia. 

At the end of the Lower Pleistocene, the large mammalian fauna first 
showed noticeable effects of Paleolithic man's presence in the environment. 
These effects were more pronounced in the western part of the Isthmus. 

Commencing in Upper Pliocene time, the eastern Transcaucasian 
landscape, composed of dry foothills, juniper -pistachio forests and steppe 
grasses dry in summer, persisted throughout the Middle Pleistocene. The 
faunal complex of this region had a mixed composition of animals from the 
Pleistocene steppes of the Russian Plain (corsac fox, saiga, etc. ), from 
166 the southwest Asian uplands (jungle cat, cheetah, mole vole, etc. ), 

and from deserts of the Turan type (small five -toed jerboa). It also included 
endemic Caucasian species (Apsheron vole). 

As a general rule, cave bear, elephant and bison were not to be found. 
The only northern steppe species from the Pleistocene Russian Plain to 
occur in eastern Transcaucasia during the Middle Pleistocene were corsac 
fox, great jerboa and saiga (Figure 76). 



160 




Caucasian mountain- 
forest and mountain- 
meadow complex 
Southwest Asian 
mountain-steppe complex 



South Russian steppe 

complex 

Central Asian semidesert 

complex 



FIGURE 76 



The ranges of the "northern" and "southern" steppe species expanded 
along the margins of the Isthmus toward each other, probably moving closer 
together in irregular waves —irregular, that is, in the sense of both space 
and time. It seems possible that the extension over the Caucasus of the 
ranges of some other northern steppe species (e.g. , common vole. Bos 
primigenius) also proceeded unevenly even in the Middle Pleistocene. 
The process was controlled by the alternation of arid and humid climatic 
phases. 

The proportion of faunal elements from the Russian Plain steadily grew 
throughout the Pleistocene. The effects of this incursion of Russian Plain 
species was miore pronounced in the fauna of the Ciscaucasian plains than 
in that of the Transcaucasian. 

True mammoths and bison, but of somewhat diminished size, lived in the 
western Ciscaucasian foothills at the beginning of the Upper Pleistocene, 
at a later time than that of the Binagady complex. 



161 



167 The extensive development of steppes on the Trans-Kuban Plain at the 
beginning of the Upper Pleistocene can be seen in the occurrences of grass 
beetle, horse, ass and saiga remains at the Il'skaya site. Mammals from 
the Upper Pleistocene localities of the Ciscaucasian plains also indicate 
the degree of development of the steppes. The mammals of this region 
include horse, boar, red deer, saiga, bison and Bos pri mi genius. 
Such an assemblage may be characteristic of either a steppe or a forest- 
steppe, but not of extensive forests nor of mesophilous meadows, and 
certainly not of moss swamps and tundra. 

From available paleofaunal data, it appears that glaciations in Eastern 
Europe and on the Caucasian mountains did not noticeably affect the 
composition of the faunal complexes on the piedmont plains. Moreover, 
development of desert and semidesert landscapes, rather than mesophytic 
landscapes, is indicated by the large collections of bones from the shoals 
of the lower Don. The occurrence of Upper Pleistocene and Holocene bones 
of blue hare and reindeer along the lower Don must be regarded as 
evidence of a late extension of their ranges southward along the valleys of 
the larger rivers of the Russian Plain, and not as a direct shift in the 
distribution areas caused by the cooling and glaciation of northern Europe. 
Mountain forms, such as snow vole, Caucasian goat and chamois, are 
completely unknown in the Pleistocene of the Caucasian foothills and plains. 
This might be taken as a contrary indication to the assumptions of 
continental glaciations and major shifts in the phj^olandscape (such as 
displacement of alpine groupings to the plains) held by earlier zoologists 
and contemporary geomorphologists . 

If, during the glaciations of the Caucasian mountains, the glaciers 
descended to the piedmont plains, then the highland fauna (mountain goat, 
chamois, Prometheomys) must have either been completely displaced 
to the piedmont plains, or migrated to some adjacent mountain refuge, 
or become extinct. 

The cooling and the concomitant down-slope movement of the altitudinal 
zones can be traced through paleontological material which reflects the 
wider distribution of mesophilous animals and animals of the mountain- 
forest zone from the time of the Middle Pleistocene in western and eastern 
Transcaucasia. 

It should be noted that the Caucasian Isthmus, because of its more 
southerly location and more varied relief, had a more variegated species 
population during the Pleistocene than the southern part of the Russian Plain. 
Evidently, during the Pleistocene a number of subtropical and hydrophilous 
species of mammals found refuge in Transcaucasia. A number of such 
forms, such as macaca, black rat and porcupine, survived through the 
Pleistocene in Colchis and Asterabad because of proximity to the Black Sea 
and Caspian basins. 

All of this is fairly well in agreement with the results of paleontological 
studies in Palestine, Syria and Lebanon which are discussed in more detail 
below. Comparison of the Upper Pleistocene Caucasian faunal complexes 
with those of Syria- Lebanon indicates that the southward migration to the 
Caucasus of European forest forms (such as pine marten) followed 
the Aegean route (over the Balkans and the marginal ridges of Asia 
Minor) more frequently than the route over the plains of southern 
Russia and Ciscaucasia. 



162 



168 During the Pleistocene the Caucasian Isthmus was an ecological barrier 
to the dispersion routes of the Russian Plain and southwest Asian species. 
The barrier prevented the southward migration to Transcaucasia of a number 
of steppe mammals of the Russian Plain, such as Siberian polecat and 
suslik. The barrier also blocked the migration of animals of the upland 
steppes of southwest Asia (southern-type gerbils, mole vole and other 
species) to Ciscaucasia. 

The Recent fauna does not reflect a series of xerothermic phases in the 
Pleistocene . 

At present one can speak of only one phase of development of the steppes 
in the Middle Pleistocene, the postglacial dry climatic phase. 

The warm (interglacial? ) phases during the Pleistocene did not bring a 
return of the subtropical fauna or flora from south and southwest Asia and Africa, 
even to Transcaucasia. The fauna of the foothills and low plateaus remained 
xerophilous and relatively thermophilous in character. Some species of the 
Asia Minor and Iranian uplands moved north by way of Transcaucasia, and 
animals of the Turan deserts migrated south and west through Ciscaucasia 
mostly through the Manych area, but there were no new immigrations of 
southern species into Transcaucasia. 

The patterns and stages in the development of the Pleistocene Caucasian 
fauna are better understood when the fauna is compared with the Pleistocene 
faunas of the adjacent areas: the Russian Plain, the Crimea and southwest 
Asia. Comparative studies can be done on the Middle and Upper Pleistocene 
faunas; the data for the Lower Pleistocene and for the Pliocene -Pleistocene 
transitional period are insufficient. 

The total number of known species of mammals in the Middle and Upper 
Pleistocene of the Russian Plain is 74. (The number is based on the 
identifications of Gromov (1948), Pidoplichko (1954) and our unpublished 
data for the Volga and Don areas). The faunas of the Russian Plain and of 
the Caucasus have 36 species in common, which amounts to nearly 50%. 
The quantitative faunal resemblance might be even higher if the alluvium 
in the river valleys of the Ciscaucasian plains were to be successfully 
searched for fossils. 

The previously noted absence of boreal species (arctic fox, blue hare, 
lemming, reindeer and musk-ox) and the presence of mountain species 
(goat, chamois, sheep and a number of southwest Asian upland rodents) in 
the Caucasian fauna of the Pleistocene are the main features that distinguish 
it from the fauna of the Russian Plain of the same period. 

Of the 58 species of Pleistocene mammals known in the Crimea (Birulya, 
1930a, b; Vinogradov, 1937b; Gromova, 1935a; Gromova and Gromov, 
1937; Gromov, 1948), 35 species (approximately 50%) also occur in the 
Caucasus. Among the species in common are the following representatives 
of plain and steppe assemblages: long-eared hedgehog, wolf, fox, corsac 
fox, cave hyena, badger, Panthera leo, great jerboa, small five-toed 
jerboa and mammoth. 

The main difference between the faunas of the two areas is the absence 
in the Caucasus of steppe species (small pika, large-toothed suslik, 
Eversmann's hamster and yellow steppe lemming) and of northern species 
(arctic fox, blue hare and reindeer) which are highly characteristic of the 
Upper Paleolithic of the Crimea. The pronounced resemblance in other respects 
of the large-mammal faunas of the Crimea and the Caucasus may indicate 
169 that the Pliocene -Pleistocene faunas in both regions evolved from a common 



163 



stock, and even that the regions were connected in the Lower Pleistocene 
by a landmass in the area presently covered by the Sea of Azov. It was the 
fauna of the southern part of the Russian Plain which primarily contributed 
to the faunas of the Crimea and the Caucasus. 

Any discussion of the faunal relationships in the Pleistocene between 
the Caucasus and the areas to the south and southwest must be based on 
a sound paleontological record of the southern Black Sea coast and inner 
Anatolia. However, there are no data on the Pleistocene faunas of those 
regions, although the activity of Acheulean man has been traced in the 
interior of the country by Pittard (1929), Pfannenstiel (1941) and Sauter 
(1948). 

According to Sauter, the interior of Asia Minor (Lake Tuz-Golii and 
other regions) was inhabited by prehistoric man from the early stages 
(Chellean-Clactonian- Acheulean) to the Mesolithic. This indication of 
natural conditions favoring the maintenance of human life means that large 
mammals were abundant in the Quaternary in regions bordering the Caucasus 
which are today deserts and salinas. 

Furon (1955) has noted that inner Turkey was less of a desert in the 
Pleistocene than it is in the Recent. Lakes Tuz-Golii and Budur-Golli were 
100 m deeper than their present depth. Of the Pleistocene cold phases, only 
the Wiirm phase is recognizable. 

The Pleistocene fauna of the southern parts of southwest Asia and 
northeastern Africa is known mostly from the studies of Blanckenhorn 
(1901, 1910, 1921-1922), Vaufrey (l93l). Bate (1937), and Picard (1937). 
Of the 69 species reliably identified by these authors, 22 (approximately 
3 1 %) also occur in the Caucasus . If species which are in doubt (Panthera 
leo. Bos and woolly rhinoceros) are included, the resemblance is even 
greater. As Picard correctly noted, there is no evidence that elk and 
reindeer lived in Syria or Palestine in the Pleistocene. Nor did they live 
in Egypt, contrary to Berg's uncritical quotation (1947, p. 82) of 
Blanckenhorn's statement on the occurrence of these northern forms. In 
the study of the development of the Caucasian fauna, it is important to take 
into account the occurrences of pine marten, European wildcat, Asia 
Minor hamster, tiger polecat, roe deer and other forms in the 
Pleistocene of Palestine. The late, postglacial stage of infiltration of 
southern Asian species into the Caucasus is represented by occurrences 
of jackal, striped hyena, jungle cat, kulan and goitered gazelle in the Upper 
Pleistocene. The core of the Pleistocene fauna of Syria- Palestine are the 
southwest Asian and north African (Mediterranean) forms, such 
as civet cat, lion, true ass, warthog, hippopotamus and Sinai ibex. The 
distribution ranges of these species in the Pleistocene probably did not 
include areas near the Caucasus. The confirmed absence of mammoth and 
the presence of the Acheulean Elephas trogontherii are important 
indicators in tracing the extinction of some "northern" species and the 
southern range extension of others (via the Aegean landmass and the 
Caucasian Isthmus) in the Pleistocene. 

The paleontological material of Palestine has been studied in more detail 
with reference to evolution of climates and landscapes than has the material 
of the Caucasus. Bate (1937) attempted to represent the climatic changes 
graphically. Taking the bone -count as an index, he used several gazelle 
170 species as representative of a xeromorphic landscape, and the fallow deer 



164 



EPOCHS 

AND 
STAGES 



Neolithic and 
Mesolithic 



Aurignacian 



Mousterian 



В 



Mousterian 



Acheulean £ 



Acheulean 



Percentage and total numbers of bone fragments 



m so 80 70 60 so 40 30 20 10 
10 20 30 ^ 50 60 70 80 90 100 




QazeUa 



*B,3S8 



Dama 
mesopotamica 




205 




100*^^^ 
-*^^ 



CLIMATE 

AND 

FAUNA 



Very dry 
and warm 



Advance of 

warmth and 

aridity 

Slight 

humidification 

Increasing 

aridity 

Warm 



Humid with 
sharp changes 
in the fauna 



Recurrence of 
dry conditions 



Warm and 

humid, with 

occasional 

aridity 

Mixed fauna 
(hippopota- 
muses and 
rhinoceroses) 



Humid, 
tropical 



FIGURE 77. Variation in number of bones of Gaze 11 a and Dama during the Anthropogene 
based on food remains in caves of Mount Carmel, Palestine (excavations of Garrod and Bate, 
1937) 



(Dama mesopotamica) as representative of broadleaf forests and of 
a more mesophytic landscape . Gazelle and fallow deer were the main food of the 
ancient Carmelites inhabiting the Wadi el Mugharah and other caves. 

Bate's study showed that Dama bones predominate over gazelle bones 
mainly in the Lower Acheulean, in the Upper Mousterian and in the Upper 
Aurignacian. According to Bate, hippopotamus and rhinoceros lived in 
Palestine at the time of the Lower Acheulean when the climate was dry 
and warm, as it was in the Middle Mousterian. The Upper Mousterian brought 
a humid, pluvial climate, with an abrupt change in fauna, and the Upper 
Aurignacian marked the last weak advance of humid conditions (Figure 77). 

The pluvial stages in Palestine probably corresponded to mountain 
glaciations of the Lesser and Greater Caucasus. Picard (1937) concluded 



165 



that statements of Blanckenhorn's (1910, 1921) and others that the 
Pleistocene climate of Palestine was similar to the present climate of 
171 Germany are erroneous. The paleontological data indicate that the clinnate 
of Palestine was of the Mediterranean type, fairly stable and dry, from the 
time of the Pliocene. 

Cooler and pluvial periods, according to Picard, are represented by 
gravels possibly corresponding to the m.ajor glaciations in the north. The 
main pluvial stage in Palestine correlates with the maximum, Riss, 
glaciation of Europe. In addition, two poorly developed pluvials occurred 
between the Acheulean and the Mesolithic. As a whole, the landscape of 
Palestine was dry desert with islands of broadleaf forests in the more 
humid valleys, that is, a landscape similar to the Recent. 

Three localities with Pleistocene mammals are known from the region 
south and southwest of the Caucasus (northwestern Iran). The mammals 
occur in diluvial sediments of marly hills near the town of Maragheh and 
in the Bisotun and Tamtama caves. A. Gebel's collection from the 
Maragheh vicinity (deposited in ZIN; see Brandt, 1870) contains fragmentary 
bones showing the Pleistocene type of preservation: Canis lupus, 
Crocuta spelaea. Rhinoceros cf. tichorhinus, Equus 
caballus, E. cf. hemionus. Bos sp. , Ovis cf. ammon. 

Paleolithic beds of the Tamtama cave, west of Lake Urmia near Rezaiyeh, 
which was excavated by Coon (l95l), yielded fifteen identified mammal 
species, e.g., gerbil (Meriones), porcupine (Hystrix), deer (Cervus 
elaphus), goitered gazelle (Gazella subgutturosa)."= Bones of deer 
constitute nearly 66%, and bones of horse nearly 12% of all the collected 
material. 

Sixteen species of mammals are known from the Upper Paleolithic and 
latest beds of the Bisotun cave, near Behistun (a rock landmark) on the 
Baghdad-Hamadan road. The finds of jackal (Canis aureus), panther 
(Panthera pardus), pika (Ochotona sp. ), goitered gazelle and other 
forms are of considerable interest. The most abundant are deer (nearly 
50% of the collected material); these are followed by horse (23%) and 
goitered gazelle (12%). A seal tooth reported from this site was probably 
introduced from the Caspian coast; beaver remains were erroneously 
recorded. 

This extremely interesting material indicates the northern range limits 
of some southern species not found at the Binagady locality. The fauna 
indicates a late Pleistocene climate and landscape in the northern part of 
the Iranian plateau quite similar to the Recent. However, the abundance 
of deer indicates well-developed tugai forests, which were later destroyed 
by man. 

The stratigraphic and geographic distribution of Pleistocene species 
in the Caucasus and adjoining areas is given in Table 62. 

The figures in Table 28 reflect the extent and detail of present knowledge 
of the Pleistocene "faunas" of the Caucasus and adjacent regions. From 
them it is possible to estimate the degree of faunal resemblance and the 
age of faunal connections between regions. They also confirm what has 
already been said of the predominantly northern and northwestern 
(southwestern [sic]) influence on the fauna of the plains and foothills of the 
172 Isthmus in the Middle and Upper Pleistocene. At that time the areas of 

* Following Fraser's identification, Coon mentions the occurrence of a European beaver (Castor fiber), 
which is evidently a misidentification of porcupine bones. 

166 



distribution of the southern (southwest Asian) and local (Caucasian) species 
either had already moved to the south, or had been reduced to isolated 
relicts. 

A survey of the stages and paths of development of Caucasian fauna by 
zoogeographic subdivisions can be made at this point only for the Upper 
Pleistocene, for which the following categories are recognizable: 

1. Ciscaucasian steppes and forest-steppes, including the Ciscaucasian 
plains, the Stavropol Plateau and the Taman Peninsula, from the Manych 
in the north to the foot of the Bolshoi Range in the south, and to the 
Apsheron Peninsula in the southeast. The fauna is characterized by 
mammoth, bison, horse, cave bear, giant deer and saiga. 



TABLE 28. Degree of similarity between Middle and Upper Pleistocene mamrnalian "faunas" of 
the Caucasus and adjacent territories* 





Caucasian 
Isthmus 


Russian Plain 


Crimea 


Southwest Asia 


Number of species known from 
Middle and Upper Pleistocene .... 

Percentage of species in common 
between Middle and Upper 
Pleistocene 


83 
100 


74 
54 


58 
53 


69 
33 







These estimates rapidly become obsolete as new localities are discovered and new collections made 
available. 



2. Caucasian mountain region, including the mountain system of the 
Greater Caucasus. In this region mesophilous mountain-forest types of 
mammals predominate: European brown and cave bear, mole vole, goat, 
chamois. Also present in the eastern part of the region are a few xerophilous 
species: Radde's hamster, wild goat and little suslik. These species 
originated in the Lower Pleistocene and have survived to the present in 
Dagestan and Kabarda. 

3. Transcaucasian foothills region, which includes the foothills and 
the plains of western and eastern Transcaucasia. The eastern section of 
the region is characterized by southwest Asian rodent, southern-type 
rhinoceros, and horse and tur; the western is characterized by bison. 

4. The region of the Lesser Caucasian plateau, characterized by an 
abundance of horse, mouflon and Bos primigenius. The hot valleys 
of the southern part of the region were probably inhabited by the camel and 
the northwestern slopes by mountain-forest species: mole vole, goat, 
chamois. 



167 



173HOLOCENE LOCALITIES 

Conditions under which animals died and formation 
of deposits in the Holocene 

The conditions under which animals died and their remains accumulated 
were generally constant from the Pleistocene into the Holocene but for some 
important changes in the later period. As the dry warm period advanced, 
there was a reduction of slope erosion by running water and consequently 
the deposition of diluvium and alluvium decreased, and fewer bones and 
animal remains were transported to river deltas, lakes and marine bays. 

Man began to affect the landscape markedly by cutting and burning trees 
and changing the flow pattern of streams and, as the result of his direct 
and indirect influence on the landscape, the large mammals of the plains 
began to decrease rapidly in the Holocene. As another consequence, natural 
deposits of animal remains became very rare. 

Some cases of mass death of Recent animals from natural causes and 
the modes of their deposition are described for the plains of the Caucasian 
Isthmus by Kolesnikov (1950) and Vereshchagin (l951d). 

Mesolithic, Neolithic, Copper-, Bronze- and Iron-Age sites with bone- 
bearing beds have been discovered (Piotrovskii, 1949) and these remains 
of human cultures of the postglacial epoch in the Caucasus have been 
thoroughly studied. The map of Paleolithic localities, shown in Figure 78, 
is after Krupnov. 
174 The scale and the location of human settlements changed radically in 

the Neolithic. Settlements were located on the shores of lakes and on river- 
banks, and cave dwellings remained only as relicts of the Paleolithic until 
they reached a new stage of development during the feudal wars of the Middle 
Ages. 

Because there was little tectonic activity in the Holocene, marine, 
deltaic and lacustrine deposits remained undisturbed, as they had duringthe 
Tertiary and the Pleistocene, and consequently they lie unexposed today. 
The possibilities of finding bones in relatively young deposits in diluvium, 
small niches and caves are considerably greater. Such accumulations of 
bones resulted from temporary settlements and predator activities, 
particularly those of owls and eagle owls. 

Bone fragments similar to the Paleolithic constitute the food remains 
in the deposits of Holocene settlements. However, many more bone and 
horn implements were found from the Holocene than from the Paleolithic. 

Occurrences of bones gnawed by domestic dogs are highly characteristic 
of the Holocene, but phalanges and metapodials from which bone marrow 
was extracted were no longer as thoroughly broken as in the Paleolithic. 

Beginning with the Eneolithic, certain bone accumulations in the Caucasus 
which contain complete skulls and skeletons of both wild and domestic 
mammals give evidence of being ritual burials. The religious -cultural 
customs of several Caucasian mountain tribes account for peculiar 
ceremonial accumulations of domestic and wild animal sku'.s in the last . 
few centuries of our era. 

The discussion which follows, organized into geomorphological regions, 
treats all these types of deposits, some of which have been studied to a 
greater and some to a lesser extent. 



168 



:i73; 



isjjj 01 риоээз mojj 




169 



V/estern Ciscaucasia and the south 
Russian Plain 



175 



The establishment in the Holocene of firm land connections between the 
Ciscauca?ian and south Russian plains enabled mammals to migrate freely 
from one region to another. 

Investigators studying Pleistocene fauna in the lower Don region 
uncovered abundant mammal remains from the Holocene. These bones 
were washed out of floodplain sediments and redeposited on sandy-gravelly 
spits. Archaeological excavations of settlements and mounds in this sam.e 
area which were carried out by the Institute of Material Culture, AN 
S.S.S.R., also yielded large collections of bones. 

Particularly valuable bone material was collected by GoI'msten during 
the excavation of mud-hut settlements on the Tsaritsa River dating from 
the fifteenth to tenth centuries B.C. , and during the excavations of the 
Khazar fortress of Belaya V'ezha (Sarkel) near the village of Tsimlyanskaya 
(Artamonov, 1952). 

Eleven species of domestic and 23 species of wild animals have been 
identified from the more than 30,000 bones collected in these excavations 
on the lower Don, mainly at Belaya Vezha. ■'■ 

The faunal complex of the steppes and plain forests is still pronounced 
in the wild mammal fauna of this period. The assemblage of game animals 
includes 16 species, ranging from corsac fox, steppe polecat and beaver 
to bear, elk and bison. The materials indicate that even in the Middle Ages 
(700 to 1300, A.D. ) the ranges of forest animals (taiga and forest-steppe 
zones) extended along the river valleys to the Ciscaucasian plains, where 
in the thickets of the floodplains of the Yeya and Kuban rivers they joined the 
Caucasian ranges of the same animals. 



TABLE 29. Mammalian species and number of bones from Early Bronze Age mud huts near Gelendzhik"* 



Species 


Number 
of bones 


Species 


Number 
of bones 


Domestic 
Horse 


2 

55 

396 

106 


Wild 
Felix lynx 


2 


Pig 


Lepus europaeus 


1 


Large cattle 


Delphinus delphis 


282 


Small cattle, primarily sheep 




2 




6 




559 


Subtotal 








Subtotal 


293 










Total 


852 









The main archaeological monuments within Ciscaucasia proper are 
Neolithic settlements, Scythian mounds and cultural layers many meters 
thick of ancient Greek colonies and of the so-called Bosphorus kingdom 
(Rostovtsev, 1918; Kallistov, 1949) on the Taman Peninsula and on the 

" This material is being published separately. 
**■ This and other similar tables were compiled from the author's unpublished data. 



170 



Kuban and Trans-Kuban plains. The Gelendzhik settlement is annong the 
oldest settlements of the region. 

In 1952 Akhanov collected many bones from pits in the Upper Neolithic- 
Lower Bronze Age mud-hut settlement on the shore of the Gelendzhik bay. 
The material consists of mottled and yellowed diaphysal and vertebral 
fragments which, when tapped, emit alight, ringing sound like calcined bones. 
Most of the bones are those of large cattle and dolphins; those left by man 
were later gnawed by some small predators. 

The wild animals are represented by deer, bear, lynx and hare (Table 29). 
Individual bones of fish, shells of marine mollusks and edible snails were 
also collected at the site. 

The wild species at the Gelendzhik settlement indicate that sea game 
was important to the ancient Tamanians, or at least that they consumed 
the bodies of dolphins washed ashore by the sea. Forest game species 
(lynx, boar, deer) were also abundant in the vicinity. 

Thick bone-bearing beds, usually no more than 2-3 m and occasionally 
as much as 10 m thick, occur on the shores of the Taman Bay, at the sites 
of the ancient towns of Cepi, Phanagoria and Taman. The beds are being 
176 eroded by the sea because of the subsidence of the coast in historical times. 
Bones of domestic animals, sturgeon-like fishes and characteristic layers 
of purple Mytilus shells comprise the food remains which are plentiful 
at these sites. Bones of wild animals rarely occur in these beds, which 
confirms the accepted concept of a high degree of cultural development in 
this region at that time (Figure 79). 




FIGURE 79, Exposures of cultural layers of Phanagoria on south shore of Taman Bay 

Photograph by author, 1952 



171 



TABLE 30. Species and number of bones of domestic mammals from Taman towns 



Species 


Cepi 


Phanagoria 


Taman 


Dog 

Cat 


4 

2 

11 

5 

15 
3 


10 

18 

6 

68 

34 

6 

4 


5 
3 


Horse 


13 


Pig 


2 


Large cattle 

Sheep 

Goat 


27 
32 


Small cattle (no closer 


identification) 


10 








Total 


49 


146 


92 







Table 30 lists bone fragments collected from several hundred meters 
of cultural layers from the three towns. 

177 From the incidence of identified domestic species it appears that the 
landscape of the Taman Peninsula in the second half of the first millennium 
B.C. , on the whole, was similar to the Recent. The nomadic herds of the 
Scythians and the local animal husbandry of the Greeks must have strongly 
depleted the vegetation cover of the peninsula at that time. Large species 
decreased and the remainder of the animal assemblage probably consisted, 
as it does in the present, of European hedgehog, fox, hare, field mouse, 
house mouse, common vole and miole rat. 

More com. plete and interesting data are available for the ancient site of 
Semibratnoe near the Cossack village of V'^arenikovskaya, east of the Kuban 
River mouth. Anfimov's excavations in 1938-1939 and in 1951-1952 shovv-ed 
that a small settlement occupied the site as early as the fifth century B.C. 
A town of considerable size developed in the third to second centuries 
B.C., whichwas later destroyed and abandoned. A small settlement of 
hunter-fishermen occupied the site in the first century A. D. The bone 
material, collected in stratigraphic sequence, completely confirms the 
conclusions drawn from study of the material culture. Particularly 
significant are the large nunnbers of bones of wild species occurring in the 
fifth century B.C. and again in the first century A. D. (Table 31). 

178 Some bird bones were also collected in the beds of this town: domestic 
chicken, wild duck and stork. The collections also include numerous bones 
of sturgeon, stellate sturgeon and cyprinid (mostly wild carp). 

Ancient animal husbandry was evidently not highly developed in the 
Bosphorus. Dogs were uniformly small, about the size of a spitz, and were 
probably used only as watchdogs. The bones of goats, sheep and swine 
are light, which indicates poor feeding. Cows were of the European short- 
hornedtype, somewhat larger than northern "peat" cattle. The occurrence 
of bones of otter, boar and deer may indicate that reed-grown swamps 
covered the area, in which there were probably no large forests. 

Bones from excavated Scythian mounds in the Maikop area, dating from 
the middle of the first millennium B.C., are known only from Vselovskii's 
drawings (1901). He depicted many tens and hundreds of skeletons of 
horses which had been killed during funeral ceremonies for tribal chieftains. 



172 



TABLE 31. Mammalian species and number of bones from the ancient town of Semibratnoe 



(177) 

Species 


Present 
Era 




Befo 


re Present Era 




CD 
3 

-a 


I 


I 


II 


III 


IV 


V 


2 ■- 
о -о 
Н .5 


Domestic 

Dog 

Horse . 


2, 2 

7/5 

15/8 

39/14 

9/6 


1 

10/6 

5/4 

13/5 

13/12 

3/2 


46/32 
102/52 

64/44 
153/67 

72/43 

17/8 


26/17 
33/18 
36/16 
87/37 
27/11 
4/3 

3/2 


25/16 
74/32 
101/42 
215/53 
98/37 
32/17 


48/31 
142/75 
103/70 
303/183 
152/72 

63/28 


99 
188 


Pig 


184 


Large cattle 


309 


Sheep 

Goat 


181 
56 


Small cattle (no closer 
identification) 


4 








25/7 
24/7 


2/1 

12/8 


1 

1 
4/2 
6/6 

1 


1 

3/3 
4/3 


2/2 

2/2 

6/6 

1 


4/4 
2/2 

4/4 

16/3 

6/5 


1 021 


Wild 


4 


Meles meles 


3 


Lutra lutra 


1 


Lepus europaeus 


7 
28 


Cervus elaphus 


35 


Capreolus capreolus 


2 


Subtotal 














80 


Total 














1,101 







Note. Slash separates number of bones from number of individuals. 



TABLE 32. Mammalian species and number of bones from mounds near Ust-Labinskaya 



Species 


Present 
Era 


Before Present Era 


3 

-a 




I-II 


I 


II 


III 


IV 


V 


VI 


2 > 
о ^ 


Dog 


1/1 

1/1 

32/5 


3/1 
19/3 

18/4 

6/2 



2/1 


17A 
5/5 

49/5 
lA 

31/2 

11/3 


90/3 
20/6 
55/6 
6A 
33/3 

13/6 


1/1 

10/1 
3/1 


1/1 


8/1 
1/1 


3 


Horse 


14 


Pig 


13 


Large cattle 

Sheep 

Goat 


18 
5 
3 


Small cattle (no 
closer identification) . . 


10 


Total 
















66 



Note. Slash separates number of bones from number of individuals. 



173 



Excavations on a limited scale carried out by Petrovskii and Anfimov 
(1937) in the towns and burial grounds of the Kuban area east of Krasnodar 
uncovered bones of horses and large and small cattle from layers of the 
fourth century B.C. to the third century A. D, 

Pieces and amulets carved from large deer antlers and bear tusks which 
were collected from these burial grounds are housed in the Krasnodar 
Museum. 

We identified only eight species of domestic animals among the 337 bones 
from the burial grounds excavated by Anfimov near the Cossack village of 
Ust-Labinskaya. The animals were sacrificed during funerals and buried 
with the dead as food supply (Table 32). 

The dogs in this region were about the size of a small husky. Swine, 
cows, sheep and goats were also quite small. 

The ancient site of Elizavetovskoe and mounds on the left bank of the Kuban, 
13 km w^est of Krasnodar, were excavated by Gorodtsov (1935, ^1936) in 
179 1934-1935. From layers of the early centuries of the present era, 

Gorodtsov recorded "many bones of cows of two varieties, small pig, sheep, 
goat (?) and three varieties of dog. Bones of wild animals are very rare; 
only antlers of red deer, placed in the grave with the dead, were found. " 

Artamonov (1937) collected remains of the following animals from Late 
Bronze Age mounds on the steppes of the northern bank of the Manych 
(near the Manych canal and the village of Spornyi): 



Domestic 


Wild 


Horse 


Lepus europaeus 


Pig 


Marmota bobac 


Large cattle 




Sheep 





TABLE 33. Mammalian species and number of bones from hills on the south bank of the K-lanych 



Species 


Xumber of bones 


Number of individuals 


Domestic 
Dog 


16 

12 

4 

71 


1 


Horse 

Large cattle 

Sheep 


2 
2 

5 


Subtotal 


103 

38 
1 

1 
2 


10 


Wild 
Vulpes vulpes 


3 


Canis lupus 




Meles meles 


1 


Lepus europaeus 


1 






Subtotal 


42 


6 






Total 


145 


16 







174 



Other species are known from the southern bank of the Manych, near 
the village of Veselyi; they occur in mounds of the middle of the first 
millennium B.C. (Table 33). Had marmot bones been found at this site, 
it would be of great paleogeographic interest, since marmots have not been 
documented on the Ciscaucasian plains by twentieth-century zoologists. 

Domestic animals were probably sacrificed in rituals. The bones of fox, 
wolf, badger and even hare, however, which were collected from the talus 
of the mound, are the remains of animals which died in their holes. 

The foothill regions and river canyons of northwestern Ciscaucasia are 
180 paleontologically unknown. Interesting data can probably be obtained from a 
study of the bones in the ancient Alani settlements, which are preserved 
in ruins in the Laba, Belaya and Zelenchuk river gorges and in the old pits 
in asphalt crusts near Khadyzhensk (Vinda, 1910). 

The only bones known from this region are the skeleton of Microtus 
(Chionomys) roberti — from a 14th-century burial chamber in the 
gorge of the Malaya Laba River, near the village of Andryukovskaya — and 
a horn of a Caucasian elk (A Ices alces caucasicus), collected in 
1949 in the gravels of the Urup River near the village of Otradnaya. 



Central Ciscaucasia 

At present the Stavropol Plateau and Pyatigor'e are peculiar meadow - 
forest "mesophytic islands" in the steppe plains. As such, they have 
retained to the present a number of mesophilous species of mammals which 
migrated to this region during the cold phases of the Pleistocene. The 
migrations came from the north, from the south and from the Greater 
Caucasus. 

Deer and lynx which have completely disappeared from the relict forests 
of the Stavropol area are known from published records (Dinnik, 1914a) 
and from material in the Stavropol Museum to have inhabited the region 
until recently. 

Small mammals are among the species which have disappeared, or are 
about to. The length of time that the little suslik was isolated in the inner 
valleys on the northern slopes of the Greater Caucasus is a particularly 
interesting paleontological problem. 

Toward its solution, bone material from the layers of the ancient towns 
has been studied and special reconnaissance excavations have been carried 
out in caves and under overhangs on the margins of the Stavropol Plateau, 
in laccoliths of Pyatigor'e and in cliffs of the river canyons cutting the 
Kabarda Plain. 

The region of the Kabarda Plain was well populated by man in the Lower 
Holocene, as shown by occurrences of Neolithic tools, numerous mounds 
and remains of ancient settlements. The Eneolithic burials at Nal'chik, 
excavated by Miller in 1929, contained ornamentations made of teeth of 
domestic and wild animals: cat, fox, boar, red deer, goat, sheep and a 
very large Bos, either domestic or wild. 

Similar ornamentations from excavations by Kruglov, Piotrovskii and 
Podgaetskii (1941) were found in the Neolithic burials of the Nal'chik mound. 



175 



near the ancient site of Agubekovskoe . The ornamentations consisted of 
teeth of boar, deer, fox, bear, and Bos. 

The Degen-Kovalevskii (1935) excavations of the 6th— 8th-century 
settlement near the village of Zeyukovo (53 km from Nal'chik) found the 
Alani food remains to consist mostly of bones of domestic animals (Table 34] 

The bull from this locality was small, as shown by its bones, and 
undoubtedly belonged to a domesticated variety. Characteristic of the 
locality is the first occurrence of domestic pig and of a very primitive 
variety of goat, with horns curved similarly to the horns of С a p r a 
pr i s с a. 

Kistyakovskii (1935) recorded remains of 15 species of extant rodents 
in the pellets of predatory birds of the steppes to the north and east 
of the Stavropol Plateau, and reported their occurrence in the same area 
of distribution as at present. 

Some bones of rodents and insectivores were collected by Lyutyi (1940) 
181 from eagle owl pellets on the western slopes of the plateau, near 

Stavropol. His data did not contain any new zoogeographical information. 



TABLE 34. Species and number of bones of mammals from Alani settlement near Zeyukovo 



Species 


Number of bones 


Number of individuals 


Domestic 
Dog 


2 

8 
15 
29 
10 


1 


Horse 


2 


Pig 


3 


Large cattle 


3 


Small cattle 









Subtotal 


64 

1 


11 


Wild 


1 






Total 


65 


12 







In the fall of 1950 we made a paleontological reconnaissance survey of 
the western part of the limestone cliff of Mount Strizhament on the 
southern margin of the Stavropol Plateau. The presence of forest, springs 
and "a sea of rocks" indicated a likelihood of finding bones of forest and 
steppe animals, left by ancient hunters and cave predators, in this area. 

In four small caves only individual bones of sheep and cows, brought 
by shepherds and wolves, were collected. The unstable ceilings and big 
blocks of limestone on the floors prevented a more detailed survey. 

A considerable number of small-mammal bones brought by eagle owls 
during the last century were collected in one of the caves on the cliff. 
The results of the identification of the bones are given in Table 35. 

The most interesting find was the remains of a pine vole, which is not 
recorded in the literature as occurring in this region, although the species 



176 



commonly occurs nowadays on the meadows of the plateau. This 
occurrence of the species, separated from its main area of distribution in 
the Caucasus, is biogeographical evidence that the climate in the past was 
more humid than in the present. 

Pyatigor'e. Studies were done on isolated elevations in the steppes 
of Pyatigor'e and compared with the Stavropol Plateau. Particular attention 
was devoted to the permafrost of Mount Razvalka, 5 km north of 
Zheleznovodsk, and to the gorge of the Berezovka rivulet south of 
Kislovodsk. 

Reconnaissance excavations were carried out on the eastern slope of 
Mount Razvalka near the Selitryanaya cave. At the site, dense broadleaf 
forest surrounds two picturesque trachytic rocks, weathered by wind 
erosion. A small dry cave, formed at the junction of three vertical joints, 
182 is located under the southern rock. Over the last millennia, deposits 
accumulated to a depth of 8-9 mi under the entrance to the cave through 
erosion of the upper cliff platforms and the rock face, and as the ancient 
dwellers cleared the cave of rock fragments. The site of the debris was 
probably used as a resting place for cattle, since the bones of small and 
large mammals and fragments of pottery occur in layers at the foot of the 
cliff and in the cone of the cave debris. 

TABLE 35. Mammalian species and number of bones from pellets of eagle owls on Mount Strizhannent 



Species 


Number of bones 


Number of individuals 




13 
25 

7 
8 
5 
3 

19 
290 
211 

1 215 

17 

30 

5 

1 

5 


2 




4 


Talpa caucasica 


1 


Mustela nivalis caucasica 


2 




1 




2 


Cricetus cricetus 


3 




34 


Cricetulus migratorius 


29 




'l 




I 32 


Spalax microphthalmus 


3 


Apodemus sylvaticus 


7 




3 


Mus musculus 


1 




1 






Total 


854 


125 







Our trial trench, 1 m wide and 3 m long, was dug 1.5-1.9 m deep 
into the talus in front of the cave entrance. The collections 
consisted mostly of simple black pottery, made with broken shells, of 
ancient Alani type. The pottery from the deeper layers is ornamented with 
pitted and striated markings, characteristic of the Bronze Age. An obsidian 
fragment, a fragment of a stone hammer and a small whetstone made of 
black slate were found in the upper 60-cm-thick layer. 



177 



183 



TABLE 36. Mammalian species and number of bones at the Selitryanaya cave in Pyatigor'e 



Species 



Depth (cm ) 



GO- 120 



Total 
bones 



Domestic 

Dog 

Horse 

Pig 

Large cattle 

Sheep 

Goat 

Small cattle (no closer identification) 

Subtotal 

Wild 

Hemiechinus auritus 

Erinaceus europaeus 

Talpa caucasica 

Vulpes vulpes 

Ursus arctos 

Mustela nivalis 

Meles meles 

Putorius eversmanni 

Vormela sarmatica 

Felis silvestris 

Lepus europaeus 

Glis glis 

Cricetus cricetus 

Mesocricetus auratus 

Cricetulus m i gra tor ius . . , . 

Arvicola terrestris 

Microtus arvalis 

M. majori 

Rattus norvegicus 

Apodemus sylvaticus 

Mus musculus 

Spalax microphthalmus. . . . 

Sus scrofa 

Cervus elaphus 

Capreolus capreolus 

Subtotal 

Total 



1 

9 

34 

103 

103 

29 

32 



311 



1 

2 

4 

2 

1 

1 

17 

10 

173 

159 

13 

31 

18 

1 

2 

2 

1 

1 

4 

1 

3 



450 



1 
8 
17 
74 
72 
5 
77 



761 



1 

4 

23 

22 

4 

10 

1 

2 



85 



2 

7 

39 



339 



123 



2 

19 

58 

216 

175 

34 

176 



680 



2 

2 

3 

1 

1 

3 

4 

2 

1 

1 

19 

15 

197 

183 

17 

41 

18 

2 

2 

4 

1 

1 

12 

5 

6 



543 



1.223 



Fragments of large -mammal bones are typically found as food remains. 
The burial also contains rare fragmentary implements made of deer antlers. 
Remains of small animals are represented by bones from the pellets 
of birds of prey, mostly eagle owls which habitually rested on the rocks 
above the cave. The locality is a rare and valuable example of a deposit 
formed by the three processes of bone accumulation: remains of food 



178 



184 



transported by birds, remains of game transported by hunters who used 
the cave for shelter, and remains of food left from ritual tribal feasts or 
transported by ancient shepherds. 

Nearly 1,500 fragments of domestic and game animal bones were 
collected in stratigraphic sequence, all showing an approximately uniform 
mode of preservation. In small fracture, they are yellow or white in color, 
and very few show any signs of fossilization. 

It can be estimated that the layer accumulated over the last 2,000- 
2,500 years. 




FIGURE 80. Cave inhabited by eagle owl on Mount Lysaya. 
On top are bodies of seven hedgehogs and one corncrake 

Photograph by author, 1952 

Six species of domestic and 26 species of wild animals have been . 
identified in the collection of bone material (Table 36). 

Remains of birds, toads, frogs and shells of Glaus ilia and Helix 
were collected at the same locality. 



179 



At the time of the ancient Alani settlement most of the Razvalka slope 
was probably deforested and man's occupation of the site probably drove 
away the eagle owls and other predators. As a consequence of their 
disappearance from the area at that time, the deeper layers of the deposit 
show a sharp decrease in the number of rodent bones. In the main, the 
deeper layers reflect man's activities, and it is only in the last quarter of 
our era, when Pyatigor'e and Razvalka again became covered with forests, 
that most of the wild-animal bones, particularly those transported by eagle 
owls, were accumulated. Large forest game inhabited the area at this later 
period: bear, boar, deer and roe deer. Old records show that deer 
disappeared from the region at the beginning of this century. Bears, roe 
deer and boars occasionally enter the forests of Pyatigor'e in our times, 
although they do not live there permanently. During the same late period, 
the last four to five centuries, the food of the eagle owl consisted of small 
species characteristic of the laccolith region in the present: three hamster 
species and water, pine and common voles. Bones of snow voles 
(Chionomys gud), hamsters and water rats were collected in rock 
crevices over the stone-covered permafrost area. 

A few specimens of snow vole were collected in the rocks of the "century- 
old freezer" of Mount Razvalka, at altitudes of 400-500 m above sea level. 
The nearest that the species has been found elsewhere is a site 3 km south 
of Kislovodsk in the gorge of the Berezovka rivulet, which is 50-55 km away 
from Mount Razvalka. Thus the occurrence of the snow vole at Mount 
185 Razvalka indicates that during a cold period in the past, probably at the time 
of maximum glaciation, the mountain animals migrated to these minor 
elevations in Pyatigor'e. 




FIGURE 81. Crevice in limestones on western wall of Berezovka rivulet gorge near Kislovodsk — day resting 
site of eagle owl 

Photograph by author, 1954 

180 



TABLE 37, Mammalian species and number of bones from pellets of eagle owls near Kislovodsk 



Species 


Number of 
individuals* 


Species 


Number of 
individuals* 




1 
3 
3 
3 
1 
1 
45 
23 
3 


Arvicola terrestrls 

Microtus arvalis 


51 




2 




M. gud 

Apodemus sylvaticus 

Mus musculus 


4 


Putorius eversmanni 


1 
2 




Rattus norvegicus 


3 


Cricetus cricetus 


Spalax m icrophtha Imus .... 


2 






Cricetulus migratorius 


Total 


148 







Number of individuals given is based on number of mandibles. 



Plentiful remains of insectivores and small carnivores in the pellets 
of eagle owls were collected in the laccolith farther northeast under the 
rocky overhangs of Mount Lysaya (Figure 80) and in the gorge of the 
Berezovka rivulet, south of Kislovodsk (Figure 81). 

186 At the latter locality the material is older and more abundant. It was 
obtained from a half-meter-thick layer of Holocene dust deposit under a 
vertical joint in the Lower Cretaceous limestone cliff. This resting place 
must have been abandoned by the eagle owl not less than a century ago, 
since it is located only 1 km from the southern outskirts of Kislovodsk. 

The bones accumulated during several centuries of the second half of our era. 

A total of 1,582 bones was collected. Their preservation is not uniform. 
The bones from the lower layer are coffee -colored with dendrites; bones 
from the upper layer are yellow. 

The species composition of the collected fauna indicates that the 
landscape at the time of accumulation of the bones was very similar to the 
present (Table 37 and Figure 82). 

Nineteen specimens of amphibians and 305 bones of birds were collected 
at the same site. The assemblage of relatively xerophilous steppe species 
is quite prominent, as at Mount Razvalka: hamster, polecat, mole rat. 

187 This assemblage never occurs in complete composition this far into the 
mountains. The moist meadows and rock-covered areas of the Berezovka 
gorge account for the large numbers of water rat and the occurrence of 
snow vole. 

The hamsters and mole rats in this region were the precursors of the 
relatively xerophilous assemblage which migrated in postglacial time from 
the dry Ciscaucasian plainsto the north. This advance overlapped by tens 
of kilometers the northernmost occurrences of the forest and alpine species 
(pine and snow voles, forest dormouse, fat dormouse) which remained 
in mesophytic sections of the laccoliths and in the river gorges after the 
recession of the colder, more humid period. The uppermost limits of 
distribution of this assemblage are the moist meadows of the Pastbishchnoi 
ridge and the lower reaches of narrow gorges cut by rivers through the 
ridge (Vereshchagin, 1953a). 



181 



;i86) 




FIGURE 82, Mammalian bones from pellets of eagle owls in the Berezovka gorge 

1 —humerus ofTalpa caucasica; 2 — skull ofMustela nivalis; 3 — pelvis of Le p us 
europaeus; 4 — jaw of Rattus norvegicus; 5, 6 — skull of С r i ce t us cricetus; 7, 8 — skull 
of M esocr ice tus auratus nigriculus; 9, 10 — skull of A г v i с ol a terrestris 



In the western part of the region, a similar picture exists, although at 
a lower altitude, on the right bank of the meridional section of the Kuban 
River. 

The following conclusions can be drawn from the study of Holocene 
animal remains in the foothills of central Ciscaucasia: 

1. The intensive exploitation of the region by man from the Neolithic 
to the present resulted in a diminution of the wooded area and the 
disappearance of large animals (bison, deer and, later, roe deer and boar). 

2. The presence of mesophilous relicts of the cold (glacial) epoch has 
been established. The relicts survived in the steppe on the Pyatigor'e 
laccoliths and on the Stavropol Plateau. This fauna includes the snow and 
pine voles. 



182 



3. A dominant assemblage of relatively xerophilous species of rodents 
(hamster, mole vole, mole rat), characteristic of open areas, existed in 
the foothills of the Kislovodsk area and further south during the last 
millennium, and possibly during the entire present era. During this period 
the little suslik did not live in the Pyatigor'e area. The migration of this 
species in the Holocene to the longitudinal valleys of the range can be 
established only after its remains have been found in layers deeper than 
those excavated. Such an occurrence might prove the existence of a 
postglacial climatic phase drier and warmer than the present climate. 



Eastern Ciscaucasia 

The steppes of the eastern Manych area are of interest to the 
paleozoologist because of the occurrence of a microlithic culture, usually 
identified as Late Neolithic. Tribal sites containing microlithic material 
occur in the southern Ukrainian and Crimean steppes, along the northern 
coast of the Caspian and in the border scarps of the Ust'-Urt, and deep 
into Soviet Central Asia. The most permanent settlements of these tribes 
were located in the Berovskie hills, surrounded by elms and rich in fish 
and boar. 

In this region, remains of material culture and fragments of animal 
bones occur in wind-blown depressions, on wind-eroded mounds and in 
sandy hills. Fragmentary bones of giant mole rat (Spalax giganteus), 
saiga and sheep occur at sites such as these in the steppes of the Kuma 
area; the great gerbil (Rhombomys opimus) occurs in blackland sands. 
Between the lower reaches of the Kuma and the Volga, the archaeqlogist Sinitsyn 
discovered numerous bones of animals of many varieties around the remains 
of ancient campfires. He also discovered many settlements and burial 
grounds of the Bronze and Middle Ages. 

Remains of a large bull with horns similar to those of В о s 
primigenius from the pre -Scythian Tri Brata mounds near Stepnoi 
were photographed by Sinitsyn (1948, p. 151). 

He dated the burials end of the second to beginning of the first millennium 
B.C. The culture, i.e., Transcaucasia, is similar to that at Trialet 
(Kuftin, 1941). 

Similar and younger monuments of the pre -Scythian and Scythian epochs 
have been traced by Krupnov (1947) in the steppes between the Terek and 
the Kuma. Osteological material from this region would greatly clarify 
the notion proposed by Krupnov that there were considerable changes in 
the landscape during the last millennium B.C. , and would probably also 
improve our concepts of the history of the ranges of central Asian desert 
animals. 

Farther south, on the northeastern slopes of the Dagestan foothills, bone 
material is known from settlements and mounds of the Bronze Age. 
Primitive archaeological excavations were carried out by Russov (1879) and 
Tsilossani (1879) in the region of Derbent, Kayakent and Deshlagar. Their 
excavation log indicates "bones of rodents, " "bones of animals, " "bones 
of large and small doinestic animals, " etc. , occurring in caves, 
mausoleums and burial mounds. The ZIN collections contain bone materials. 



183 



obtained by Kruglov ( 1946a) in excavations in 1941 from layers of a 
settlement of the second half of the second millennium B. C. near the villages 
of Kayakent and Dzhemikent. The bone fragments from these layers 
are light brown in color and they clink slightly. Bones of boars tend to 
absorb water as do Upper Pleistocene bones. Identifications and 
counts are given for these fragments in Table 38. 

TABLE 38. Mammalian species and number of bones from Bronze Age settlement near Dzhemikent village 



Species 


Number of bones 


Number of individuals 


Domestic 
Dog 


82 

129 

11 

46 

401 


1 
6 


Goat 


4 


Sheep 


5 


Small cattle (no closer identification) 


15 


Subtotal 


669 

1 
1 
1 
6 
3 
6 


31 


Wild 
Phoca caspica 


1 


Ursus arctos 


1 


Vulpes vulpes 


1 


Equus hemionus 


1 


Sus scrofa 


1 


Gazella subgutturosa 


1 






Subtotal 


18 


6 






Total 


687 


37 







189 The occurrences of seal, goitered gazelle and kulan in the region are 
noteworthy (Figure 83). 

The reconnaissance excavations of 1954 carried out under the picturesque 
cliffs of Kapchugai west of Makhachkala by the archaeologists of the Dagestan 
Branch of the AN SSSR revealed layers which have been dated late 
second to early first millennium B.C. The following bones have been 
identified by us from these layers: horses (3), large cattle (12), sheep (14) 
and red deer (Cervus elaphus) (12). These numbers indicate both the 
abundance of deer and the importance of game hunting in the economy of 
the ancient shepherd tribe. The cliffs of Chokrak sandstones at the 
site are covered with line drawings of animals, made only a few centuries 
ago. Panther, horse, deer and saiga can be recognized in some drawings. 

Rem^ains of Iron Age animals are known from excavations of ancient 
settlements in the valleys of the Sunzha River and its tributaries. 

Krupnov (1948) has published the results of excavations of a settlement 
of Scythian time (second half of the first millennium B.C.), near the village 
of Alkhaste on the left bank of the Assa River (Table 39). 

He also reports (1949) a similar ratio of domestic species 
distribution for the Assa River valley settlements of Nesterovo and Lugovoe 
(sixth to fifth centuries B.C. ; identifications by Tsalkin). At the Lugovoe 



184 



locality a horn stem of a saiga and individual bones of a roe deer have been 
identified among the numerous bones of domestic animals. 







( I i L i , L I i S, i ,.i 




FIGURE 83. Mammalian bones from the Bronze Age settlement near Dzhemikent 

1 — jaw of Vu Ipes vulpes; 2 — tarsus of U rsus arctos; 3-5 — first phalanx, hoof and astragalus 
of first phalanx and tibia of Equus hemionus; 6-9 — horn, head of femur, first phalanx and tibia 
of Gazella subgutturosa 



(190) 



TABLE 39. Mammalian species and number of bones from Scythian epoch settlement neai Alkhaste village 



Species* 


Number of bones 


Number of individuals 


Domestic 
Dog 


6 

21 

175 

101 

54 


3 


Horse 


6 


Pig 


15 


Large cattle 


10 


Small cattle (no closer identification) 


8 






Subtotal 


357 

1 
5 
2 

6 


42 


Wild 
Vulpes vulpes 


1 


Cricetus cricetus 


1 


Cervus elaphus 


1 


Capreolus capreolus 


3 






Subtotal 


14 


6 






Total 


371 


48 







' N.A. Sugrobov's identifications. Bones of bear were probably included with domestic pig. 



185 



TABLE 40, Mammalian species and number of bones from settlements and burial grounds in the 
Sunzha valley 



Species 


Number of bones 


Number of individuals 


Domestic 

Dog 


75 

5 

55 

32 

64 

145 

6 


4 


Cat 


1 




3 




10 


■^ '5 

Large cattle 


7 


Sheep 


12 




3 






Subtotal 


382 

1 

1 


40 


Wild 


1 




1 






Subtotal 


2 


О 






Total 


384 


42 







In the eastern part of the Sunzha valley, excavations of settlements and 
burials of the second to third centuries A.D. were carried out in 1938 by 
Kruglovnear the villages of Isti - Su, Alkhan-Kala and Khoro-Chai. 
The bones from these excavations, identified by the author (Table 40) are heavy 
and stained in places with humus. There are only a few remains of wild 
191 animals. However, the new evidence of past occurrences of steppe species 
of large mammals in the foothills of the eastern Caucasus is of considerable 
interest. 

In 1952 we made a reconnaissance search for species of smaller 
mammals in the cavities and talus of the Sarmatian limestones above 
Makhachkala and in the sandstone cavities near Kapchugai. Large numbers 
of bones were collected: long-eared hedgehog (Hemiechinus auritus), 
hare (Lepus europaeus), little suslik (Citellus pygmaeus), 
hamster (M e s о с r i с e t u s auratus n i gr i с u lu s) and steppe vole 
(Microtus socialis). All the bones were collected in the pellets of eagle 
owls and flying predators which settled to rest on the cliffs. 

Further paleontological studies of Holocene burials, which take into 
account the earlier archaeological data (Komarov, 1879; Shtein, 1879; 
lessen, 1935), will undoubtedly fill in the detail of the general picture we 
have of postglacial steppe species of ungulate mammals in the region. 
They will also add to an understanding of the decrease in numbers and 
deterioration of composition of the fauna toward the present. 



186 



Longitudinal valleys of the northern slope 
of the Greater Caucasian and inner Dagestan 

The longitudinal tectonic valleys, bounded in the north by the 
Pastbishchnoi and Skalistyi ridges, are well developed east of Elburz. Like 
woodless inner Dagestan, the valleys attracted post-Paleolithic tribes as 
inhabitants because of their natural shelters and dry, relatively warm 
climate. Mesolithic sites have recently been discovered in the valleys and 
in inner Dagestan. Some Bronze Age tribes left behind them remarkable 
burial monuments: mounds, underground burial sites and catacombs for 
mass burials. The sites of their settlements are represented by fairly 
thick layers of the "Kobanian culture'' (Uvarova, 1900, 1902). 

The medieval Alanis built numerous fortresses and war towers, fortified 
caves and individual trenches in the vertical, rocky cliffs. Their 
descendants, the Ossetians, built peculiar collections of hoofed-mammal 
skulls in ritual places called "dzuars. 

Many bones of rodents from the pellets of eagle owls accumulated in 
natural niches and ruins of the fortresses. 

We studied all these types of deposits, with the exception of the Kobanian 
burial grounds and settlements. 



Mesolithic sites in the Baksan and Avar Koisu gorges 

In 1955, Zamyatnin discovered and studied diluvial sediments, many 
meters thick, containing Mesolithic flint tools, on the right slope of the 
Baksan ravine at the head of the first gorge near the village of Byllym 
(Figure 84). Excavations to a depth of 6.85 m at the Sosruko overhang 
revealed a sequence of Middle Ages, Early Iron Age and five Mesolithic 
layers with characteristic flint tools of the transitional epoch. 

Most of the bones taken from every layer were broken into small 
fragments. The loss of organic matter in the bones is in its initial stage. 
Identifications of the collected bone fragments are given in Table 41 and 
Figure 85. 
193 In addition to the species listed, about ten bones of partridge, short- 
eared owl, griffon, vulture and small P a s s e r i f о r m e s were collected 
from the Mesolithic layers. 

The list of animal species, although variegated, presents no 
zoogeographic novelties. Most of the species are characteristic of the 
region in the last century, prior to the extermination of deer, chamois and 
goat on the Skalistyi ridge. The find of suslik bones in the Mesolithic beds 
is particularly interesting as related to the question of the age of the 
populations of this species in the mountains of Caucasus. However, these 
bones may have originated in the steppe plains of the Manych area and been 
introduced into the layers from pellets of eagle owls and predators flying 
south. 

The Mesolithic site of Chokh in inner Dagestan was discovered 
in 1955 by Kotovich 2 km north of the village of Chokh in the Gunib area. 
194 The site is located under a steep limestone rock (Figure 86) on the right 
bank of a small stream, which cuts deeply into a mountain ravine covered 



187 



:i92: 




FIGURE 84. Sosruko grotto in Baksan gorge 

Photograph by S.N. Zamyatnin, 1956 



1704 



188 



by an alpine meadow. The thickness of the Mesolithic layer in the excavated 
areaexceeds 0.5 m. The cultural layers contain flint tools in all stages of 
finishing and finely-fractured bones. Among the flint implements are tools 
with blunt backs, points, knifelike blades [or flakes] and various microliths. 

193) TABLE 41. Mammalian species and number of bones in Mesolithic beds of the Sosruko grotto in 
Baksan gorge 











Mesolithic beds 






Middle 

Ages 


Early 
Iron Age 










Species 


1 


2 


3 


4 


5 


Domestic 
















Pig 


7 
20 


4 
15 


7 


- 


- 


- 




Large cattle 


— 


Small cattle 


23 


28 


44 


- 


- 


— 


- 


Subtotal 


50 


47 


51 








_ 


Wild 




Vulpes vulpes 


1 
1 


- 


- 


- 


- 


1 




Martes sp 


- 


Meles meles 






1 


_ 


— 


_ 


1 


Panthera pardus 


1 


Lepus europaeus 


- 


1 




3 


2 


— 


— 


Citellus pygmaeus 


- 


- 


1 


- 


- 


1 


1 


Apodemus s у 1 v a t i с u s . . . . 


- 


- 


- 


1 


- 


- 


- 


Ellobius talpinus 


— 


— 


1 


1 


1 


— 


— 


Microtus gud 


- 


- 


- 


- 


1 


- 


- 


Sus scrofa 


— 


4 


6 


7 


3 


1 


_ 


Capreolus capreolus .... 


1 


Cervus elaphus 


1 


1 


1 


2 


12 


7 





Rupicapra rupicapra .... 


1 


Capra cf. caucasica 


— 


- 


4 


1 


11 


- 


— 


















identification) 


7 


3 


73 


125 


526 


198 


45 






Subtotal 


10 


9 


87 


140 


556 


208 


50 






Total 


60 


56 


138 


140 


556 


208 


50 







The state of preservation of the fragmented bone material is similar 
to that of fragments from beds of the third and second millennia B.C. in the 
foothills of Dagestan. Identifications of the bone material in the 1955-1957 
collections are given in Table 42 and Figure 87; most of the fragments are 
unidentifiable. The rodents are represented by species which are 
characteristic of the region in the present. They were introduced into the 
site from the pellets of predatory birds. Among the ungulate bones, the most 
interesting are those of a bison and a small ram, probably a successor to 
the rams of the Lower Paleolithic Kudaro site. 



189 




FIGURE 85. Mammalian bones from the Mesolithic layers of Sosruko grotto 

1 — jaw of Pan thera pardus; 2 — jaw (x 2) of С i t e llu s py gm a e us; 3 — jaw (x 2) of 
Ellobius talpinus; 4 — jaw of С e r vu s elaphus; 5 — astragalus of С a p ra caucasica 



From the Koban burial grounds of the Bronze Age in Ossetia, 
archaeologists long ago collected sculptures of local domestic animals. 
These sculptures were made of bronze or, more rarely, of gold and silver. 

Pendants, fibulae, metal plates, buckles and heads of staves were often made 
by Koban craftsmen in the forms of heads of Caucasian goats, deer, elks, 
wild and domestic rams, bulls, goats, horses and bears (Uvarova, 1900). 

The past occurrence in the Caucasus of some of these species was 
confirmed by our studies of Ossetian burials. The materials from those 
sites indicate a late stage of existence of a rich assemblage of mammals 
on the northern slope of the Greater Caucasus. 



195 Dzuars of North Ossetia 

Collections of skulls of domestic and wild ungulates from ritual sites 
in the Caucasus have long been known. These collections were mentioned 
by 17th century European travellers in Circassia (Jean de Luc, 1879). 
Reference is made to them in the notes of Guldenstaedt at the end of the 18th 
century (Pallas, 1831), and in papers by Dinnik (1890a) and Satunin (1914). 
The period of accumulation of skulls in the dzuars extends over the last 
five centuries. 



190 



TABLE 42. Mammalian species and number of bones from the Mesolithic site at Chokh in Dagestan 



Species 


Number 
of bones 


Species 


Number 
of bones 


Domestic (?) 
Horse 


1 

1 
1 
5 
5 
1 
48 
302 


Capra sp. (cf. aegagrus or 
C. cylindricornis) 

Ovis or Capra wild 

Caprovinae (no closer 
identification) 

Bison bonasus (cauca sicus ? ) 

Small fragments of skeletal bones, 
mainly Artiodactyla 


8 


Wild 
Felis lynx 


94 

7,202 
20 


Lepus europaeus 




Mesocricetus auratus 

Arvicola terrestis 


712 


Sus scrofa 


Total 


8,400 


Cervus elaphus 




Ovis sp. (cf. gmelini) 





196 



In 1947 and 1948 we studied 18 dzuars in the ravines of Urukh, Ardon, 
Fiagdon and Gizel'don (Figure 88). 

Skulls from dzuars located on the piedmont plains, e.g., the sacred 
Khetag grove, were taken in the thirties to bone -calcination plants. Up to 
30% of the dzuars in the mountain zone were also emptied of their bones 
(Vereshchagin and Semenov-Tyan-Shanskii, 1948; Vereshchagin and Naniev, 
1949). 

The Adygeians, Digorians and Ossetians built their ritual sites in 
picturesque groves on the spurs of ridges, under rocky overhangs and in caves. 
Special buildings were also constructed: churches, chapels or large 
enclosed sheds made of stone or of large logs. 

Dzuars were dedicated to different deities and each one belonged to a 
particular village; the villagers assembled there for feasts on special annual 
holidays of spring, summer or fall. 

The custom of killing animals in the dzuars during feasts survived from 
pagan times in Abkhazia. According to Abkhazian legend (Veidenbaum, 
1879), a white bull used to emerge from the Oggin cave and was sacrificed. 

In addition to ritualistic killing of domestic hoofed mammals, the 
Digorians and Ossetians used to bring skulls of wild, hoofed animals to the 
sites. This custom of bringing skulls into the dzuars was connected with 
the belief in the god Avsati, protector of hunters and game, to whom the 
skulls were dedicated. The occurrence of bison leg bones, with hoofs intact, 
in the deep skull-bearing beds of the Digorized cave suggests that special 
hunting expeditions set out before the holidays to bring an intact deer, elk 
or bison into the dzuar. Skulls of wild, hoofed animals, boiled by hunters, 
were also brought into the dzuars without relation to holidays. They were 
either carefully laid in one common pile or on special shelves and supports, 
or they were hung on cut branches of trees and placed against the walls of 
the dzuar. It is difficult to estimate the age of the collections from the 
number of wild animal skulls. 

The state of preservation was not uniform among the skulls of wild 
hoofed animals from different dzuars, because of differences in treatment 



191 



197 and preparation. No skulls had intact lower jaws; of the bison skulls, only 
the hardest frontal and occipital bones with horn stems attached, were 
preserved. This mode of preservation is identical to that of the skulls of 
primitive bison from the shoals of the Don, the Volga and the Ural. 




FIGURE 86. Chokh site on the Gunib 



Photograph by V. G. Kotovich 





FIGURE 87. Mammalian bones from Mesolithic beds of 
the Chokh site 

1-3 — First phalanx, metatarsal epiphysis and astragalus 
of О vis cf. gme lin i 



Only a few skulls were preserved in their entirety with basal part, 
intermaxillaries and horns intact. The horns were fastened to the stems 
with iron arrowheads inserted into holes prepared for this purpose. Horns 
were also removed from their stems and used as wine goblets. Boiled 
skulls of turs, chamois and roe deer were usually brought in after the basal 



192 







FIGURE 88, Map of some dzuars in North Ossetia 

1 — remaining forests; 2 — ritual sites studied; 3 — 1 imits of distribution of deer in the 19th century; 
4— area inhabited by relict bison and elk at the end of the 18th century; 5 - Recent glaciers 



193 



part had been removed with an oblique cut. The horns were removed, 
together with skin and parts of frontal bone, from the fresh skulls of male 
deer and elks killed in the fall. The brain was then removed through the 
198 skull opening. The skulls were boiled, cleaned of flesh and small bones 
and placed in the dzuar. All the skulls of female deer and elks are without 
occipital bones which were excised to the dorsal mid-height. Most of the 
deer skulls did not have nasal parts, as they separated from the skull in 
the boiling process. 

The most remarkable dzuars are Digorized (Olisai-Don), Lesgor I and 
Lesgor II and Rekom. 




FIGURE 89. Digorized cave in Jurassic slate cliff in Urukh gorge 

Photograph by author, 1947 



Digorized cave is located on the right bank of the Urukh gorge in 
the southern cliff of the Skalistyi ridge. The cave was formed in the process 
of the weathering of Jurassic limestones. Ledges, picturesquely covered 
by pines, juniper and dog rose, occur on the giant cliff. 

The cave is approximately 1,200 m above sea level. It lies within the 
village limits of Zadelesk, located 1.5 km to the south. 

The mouth of the cave is triangular in shape, approximately 12 m high 
and 5.5 m at the base. The entrance is covered by a strong wall, 2.5 m high, 
made of limestone fragments (Figures 89 and 90). 

The feasting hall is approximately 7 m wide and 12 m long and contains 
a primitive fireplace, narrow benches and tables made of split pine logs. 



194 



an iron cup for donations, two knotty-wood trunks placed against the wall, 
carved wood fetishes hanging on the trunks and some utensils, consisting 
of two large meat bowls, a wooden spoon and about fifteen horns of bison, 
used as goblets. A pile of skulls and horns in the far left corner is heaped 
199 between two fallen limestone blocks. A support, 2.7 m high, built of logs 
and boards, stands in the forepart of the hall. Some 140-150 pairs of 
antlers are piled on the support (Figure 134). The walls of the hall are 
covered with resinous soot from the fireplace. The entrance to the narrow 
right branch of the cave is blocked by an upright wooden log with a tin icon 
of St. Nicholas. At the entrance, the height of the branching chamber is 
about the height of a man; the height decreases and the chamber ends eight 
meters from its entrance. The floor is covered by a meter-thick layer of 
skulls and horns of artiodactyls (Figure 91). 

The annual feasts in the Digorized cave took place in June during the 
mowing season. 




FIGURE 90. Map of Digorized cave 

1 — fallen blocks; 2 — fireplace; 3 — ceremonial pottery; 4 — benches; 
5 — support with deer antlers; 6 — stone wall; 7 — skulls of domestic 
goats, roe deer; e — elk skulls; b — bison skulls 



195 



On our first visit to the cave in August 1947, three elk skulls and 
15 bison skulls were found in the front rows of skulls on the floor of the 
branch. These were probably the skulls noted by Guldenstaedt and Dinnik. 
Dinnik mentioned three camel- skulls which were not in the cave; he probably 
misidentified the elk skulls. The pile of skulls in the left corner of the 
feasting hall was covered by centuries -old dust layers, whereas the skulls 
in the right branch of the cave had evidently been recently disturbed by 
archaeologists or tourists. As the layer of skulls was being sorted, bent 
or looped twigs (symbolic yoke), planed, half-burned sticks (shashlik spits), 
iron arrowheads (Figure 92) and small silver coins were found. 

The greatest number of bones found at Digorized were skulls of sheep, 
and, in descending order, skulls of deer, large cattle, domestic goats, 
bison, roe deer, chamois, elks, Caucasian turs, wild boars and buffalo. 
The ratio was established by the value placed on the hunters' sacrifices 
and by the availability of various game species in the hunting area around 
the village of Zadelesk. The bison and elk skulls are the best-preserved 
bone material because they remained on the floor of the dry parts of the 
cave, between the entrance and the side branch. In many cases, ligaments 
on the anterior and posterior sections were preserved because of the 
absence of hide beetles. On the frontal bone of one of the bison skulls the 
figure "1833" had been engraved with a knife. The elk skulls were in the 
same state of preservation. Remains of tree twigs and bushes were well 

200 preserved in the crevices of the animals' teeth. No elk horns were found 
in the cave. The horns had been removed from the frontal and occipital 
parts of one of the elk skulls in the front row . In another skull, found in a far 
corner, the horns had either fallen or been broken off the crown. It is 
possible that the horns were stolen from the cave or retained by the hunter 
as a particularly valuable trophy. 

Deer antlers, found mixed with skulls in places protected from rain 
and snow, retained their natural brown color. Traces of molting were 
preserved on only a few antlers of deer which had been killed in August. 
Most of the bones left on the support, however, became uniformly whitened 
on the upper surface by rain, snow and sun, and blackened underneath 
by soot. Skulls of chamois and roe deer occur mainly in the middle and 
far section of the bone layer. Their state of preservation was much the 
same as that of the bison and elk skulls. 

The Dagestan goat is represented by four skulls with preserved horns; 
their condition indicates that they were middle-aged. The three skulls of 
wild boar found in the upper layers piled to the side can probably be 
accounted for by the fact that the Digorians in later times abandoned their 
earlier religious prejudices against this animal, and also by the fact that 
the supply of more valuable animals dwindled. Fresh-boiled skulls and 
limb bones, fractured by an ax, and ribs were only those of large cattle, 
goats and sheep. It was clearly observed that large skulls of ancient, 
mature bulls had been replaced in recent time by skulls of small, immature 
specimens, which had been broken longitudinally. 

201 The skulls indicate that the large cattle of the region belonged to only 
one variety during the entire period of accumulation of the bone bed. These 
were large animals, with a concave area between the horns, which extended 
forward, sideward and upward. The cattle belonged to the secondarily 
enlarged European Bos brachyceros. Skulls of goats and sheep 



196 



;2oo) 




FIGURE 91. Interior of Digorized cave 



Photograph by author, 1947 




horizontal 
-L- 1 l_ 



FIGURE 92. Iron arrowheads from 
Digorized cave 



represent relatively small varieties with 
poorly developed horn stems and horns. 
If it is assumed that, from the beginning 
of the annual ritual, one bull and two to 
three sheep were sacrificed each year, 
it follows that the collection accumulated 
over not less than 500 years. 

The skulls in the right branch of the 
cave were deposited over a thin layer of 
limestone debris and old dust, underlain 
by a solid plate. A reconnaissance trench 
dug in the left corner of the feasting hall 
down to the limestone floor (a depth of 
60 cm) showed that there are no bones 
in the layer of old dust. 

Because skulls of bison and elks were 
particularly valued by the Digorians as 
ritualistic offerings to Avsati, it would 
seem likely that most of the skulls of 
animals killed by the Zadelesk hunters 
were deposited in the cave. 

It is clear that the incidence of deer and, 
particularly, of elk was very low even 
as early as the 15th to 16th centuries. In 
all probability, the period of accumulation 
of the skulls could not have begun before the 



197 



14th or 15th century and must have ended in the 18th or early 19th century. 
The accumulation in the cave of deer skulls and horns, as indicated by their 
state of preservation, ended only a little later than that of the skulls and 
horns of elk and bison. However, the deer survived until the time 
when the Ossetians obtained rifles. The Zadelesk hunters killed on an 
average up to ten deer each year. 

The Digorized bone material is particularly interesting for analysis of 
the age and sex groups of the game, as well as the seasonal hunting yields. 
These data are given in Table 43. * 

TABLE 43. Sex and age groups of wild ungulates from Digorized cave 





Females 






Males 




Species 






























immature 


mature 


total 


young 


immature 


mature 


total 


Sus scrofa 


_ 


_ 


_ 


_ 


_ 


3 


3 


Capra cy lindr icorn is .... 


- 


1 


1 


- 


- 


3 


3 


Rupicapra rupicapra 
















caucasica 


9 


15 
26 


15 
35 


1 


8 


10 
21 


10 


Bison bonasus caucasicus 


30 


Capreolus capreolus 


10 


10 


20 


- 


2 


32 


34 


Cervus elaphus maral.... 


10 


266 


275 


- 


55 


384 


439 


Alces alces caucasicus... 


— 


4 


4 


— 


— 


2 


2 



The main game animal, deer, was hunted mostly in the fall, during 
the rut. 

Dzuar Lesgor I (lower) was built on a small ledge of the same 
Skalistyi cuesta and at the same altitude as the Digorized cave, but on the 
left wall of the Urukh gorge. It belonged to the village of L^esgor, which 
was. abandoned at the beginning of this century. The skulls were arranged 
in two layers behind a stone fence on an open wall. The lower layer rests 
directly on the ground, the upper on a big pine log inserted into a crevice 
in the rock, and on a "hanger" made of a knotty-pine trunk. The upper pile 
202 of skulls is colored by the smoke of the ritual fire, which was regularly lit 
under the rock. Most of the skulls and antlers of deer in the lower layer 
decayed because of moisture. The following kinds of skulls (given in 
descending order of occurrence) were found: sheep, cow, goat, deer, east 
Caucasian goat, roe deer, bison and elk. Among them was a completely 
"fresh" skull of a young bison and a completely smoked skull of a she-bison 
with right horn intact. 

Lesgor II (upper) is located 250 m higher on the cliff on the side of 
a very steep path in a picturesque crevice. The collection of skulls is 
under a small rock overhang and is poorly protected from weather. Deer 
antlers rest on a horizontal, elevated support made of pine logs. Some of 
the skulls on the ground have "grown" into the soil and become covered 
with sod. Skulls of wild, hoofed mammals are more numerous than at 
Lesgor I, probably because the dzuar is located closer to the hunting grounds 



Only the skulls whose age and sex were identified with certainty were counted. 



198 



203 



and farther away from the village. The descending order of occurrence 
is as follows: sheep, east Caucasian goat, deer, cow, goat, bison, 
chamois, elk and roe deer. 

As at Lesgor I, the horns are missing on most of the goat skulls; they 
were broken off and probably taken to the village (Figure 93). 

This group of the Urukh dzuars is very important in the study of the 
past ranges of deer and elk and their gradual shrinkage. Since it was 
difficult to carry the skulls and horns by mountain paths to the dzuars, 
hunters chose the most venerated dzuars and those which were closest to 
their hunting grounds. From the location of the ancient mountain paths 
through the passes of the Skalistyi ridge, which connected the villages 
Zadelesk and Lesgor with the piedmont plain, and from local legend, it can 
be established that most of the skulls of deer, elk and bison originated 
below the dzuars in the forests of the Chernye gorge and on the sloping 
Terek plain, particularly in the vicinity of the village of Akhsarisar. Even 
in the 'twenties, deer still appeared during the mating season in the upper 
Urukh gorge, near the villages of Styr-Digor, Nogkau, Kamunt and Dunta. 
The last of the bison and elk, however, mainly inhabited the foothill forests 
of Ossetia toward the end of the 18th century. This is confirmed by the 
absence of skulls of these forms in the upper dzuars. In the longitudinal 
valley proper, between the Skalistyi and Bokovoi ridges, deer, elk and 
bison were probably either exterminated or driven out before the time of 
the Digorians by dense settlement and destruction of forests by earlier 
inhabitants. There are reasons to believe that elk and bison still frequented 
the forests of Ossetia at the time of Guldenstaedt's travels, although only 
sporadically. 




FIGURE 93. Skulls of east Caucasian goats in Lesgor II dzuar 

Photograph by author, 1947 



199 



Rekom Dzuar is the best known dzuar in the literature. It is 
located in the gorge of the Tseya (a tributary of the Ardon) in the pine- 
forest zone of the Bokovoi ridge and is greatly venerated by the Ossetians 
(Vyazigin, 1929). The dzuar -belongs to the village of Tsei, but is visited 
by villagers from other places as well. 

The dzuar is a massive structure dating from the end of the 17th century. 
It is constructed of pine and yew logs with one door and no windows and is 
situated in a pleasant meadow surrounded by old pine forest and rocks, 
with an outlook to snow-covered peaks (Figure 94). The site was inhabited 
as early as the Bronze Age. 

Those skulls which remain after they were plundered by tourists for 
souvenirs rest on specially constructed wall ledges and shelving of raised 
earth. The material consists of skulls and horns of Caucasian goats, 
domestic sheep, cows, goats and deer. (Satunin (1914b) reported bison 
skulls from the Rekom dzuar. ) Most of the remains of Caucasian goat 
consist of horns and soft horn stems weathered and split by moisture and 
sun. Six fresh-boiled skulls were also found, brought probably from the 
margins of the nearest glaciers during the last 20-30 years. Chamois still 
live in the rocks above the dzuar. Because of the deep winter snow deer 
and roe deer probably did not inhabit the region even in the past. Most of 
the deer skulls and horns of Rekom were brought there from the lower areas 
of the Ardon basin. 



204 




FIGURE 94. Rekom dzuar 

Photograph by author, 1947 



200 









!X 






ст 


со 


со 


ю 


го 


го 


со 










^ 


•[Pioiqns 




о о a 


J с^ 




со 


го 




о 
ю 




ю 




f- 


со 


00 








OQ ^ 


с- 




















^ 


00 




(jsddn) 


1 


о Ю <= 




1 


^ 


го 


^ 


со 




со 




со 








П joesgq 






























(J3M0t) 
































I joSsaq 










m 




t -Ч 






со 




(N 


















ю 


















































pgziJoSiQ 




■* CO o. 




со 


ю 


со 
со 






1 


ю 

со 














о о с 




























ZBUBqM 


1 


о о с 

Ю 0_ Сч 






1 


со 


1 


1 


1 


^ 




1 






ВЦПЦШП>] 


1 


со ■-( «: 






1 


(М 


1 


оо 


1 


1 




1 






с 
































« 






























































j=: 


joSiq-jAjs 


1 


с^ о U- 






1 


1 


1 


1 


1 


1 




1 






a: 
































3 






























































D 


nB>iSoN 


1 


ст> to о- 

СО Ю U- 






1 


- 


1 


(N 


1 


- 




1 






Biuna 


1 СЛ o) a 

1 C- СГ) ,- 






1 


- 


1 


- 


1 


1 




гН 








РОЦМ 


1 CO о (Г 

1 i:~ 00 u- 






1 


со 
г- 


1 


о 


1 


о 




1 








piSz 


1 rH to tc 

1 00 C- LT 






1 


1 


- 


UO 


1 


1 




1 






(laddn) 


































1 


СО 05 Г- 




























uop^s 




ю со с\ 






1 


1 


1 


^ 


1 


1 




1 








шо>]эа 


1 


го (М tr 
со с^ 






1 


^^- 


1 


го 
со 


со 


1 




1 






(jaddn) 
































jnZT^Al 


1 


со о rt 






1 


1 


1 


ю 


1 


со 




1 






































zig 


1 


■* (М ^ 






fM 


,—1 


1 


1 


1 


со 




1 






nj 






т— 1 


























л 






























































о 


(J3M0l) 




со ^ 1> 


























"Н 


§вшвлг2 


1 


со со (> 






1 


1 


1 


ю 


1 


[-^ 




1 






< 






























JT3N 


1 


ю со со 

Ю Сч1 С~ 
































^ со 


























пч>1вяв1 


1 OO lO о 

1 .-1 о cc 

^ CO 






1 


со 


1 


о 


1 


1 




1 






с 
































О с 
































■^ '.л 


siSatzq 










го 


(М 




т-Н 














.2 -° 




1 


■* со см 




1 






1 




1 


1 




1 






ь 


















































: 


3 






л 


























3 


са 




п 
































































о 


h 


о 


о 




о 














о 
















п 


о 




•^ 












<u 


■^ 












*-• 


ел 








сх ■ 




































3 ] 












(U 


E 








п] 


is 


о 


л: 
п. 




-о 
с 


о 


3 


ел 


са 


л 
























— ' 


а] ся 




3 


^ 


о 






Q 






о 




(Л 




о 


1 — 1 


<л 




с 
о 




о 














^*— 


3 




>, 




о. — 1 




*-■ 


н 






























XI 














а 


»- 


о 


(Л 


.J 




3 




XI 


efl 


3 








о 


о 


(УЗ 




(U 


3 


ел 


efl 


« 


— о 


с 


о 


со 










(U сх 




ел 


м 


> 


с\) 


м 


С' 


>- 3 


-1 
















ЕЛ 


Он 






G, 




Си rt 


















3 


л 


(U 




ПЗ 




3 о 


,^ 


о 










CO 


-J 


сг 


о 




ел 


и 


и 


<; 


и 


и 


ей 


CQ 









201 



The horns and skulls of Caucasian goats belonged to mature or even aged 
individuals of 6-17 years; the most common age is 10-11 years (Figures 
95-96). 

From the 18 dzuars we collected up to 10,000 items: skulls, horns and 
fragments of skulls and horns of 8,815 individuals. They belong to 1 1 species 
of hoofed mammals: domestic species of bull, buffalo, sheep and goat; wild 
species of boar, bison, east Caucasian goat, chamois, elk, deer and roe 
deer (Table 44). 

Our studies of the skulls in the Ossetian dzuars compared with data 
drawn from the literature and with other verified data lead to the following 
conclusions: 

1. One and a half centuries ago the Terek-Sunzha inclined plain and the 
canyons on the northern slope of the Greater Caucasus were inhabited by 
eight species of artiodactyls: boar, bison, east Caucasian and west 
Caucasian goats, chamois, red deer, elk and roe deer. 

2. Bison and elk were on the verge of extinction by the end of the 18th 
century, at which time deer were still abundant. 

3. The final extermination of deer in that section of the Greater Caucasus 
under study occurred in the second decade of the 20th century. 

4. The number of goats, chamois, roe deer and boars greatly decreased 
during the last century and the beginning of the present one. 

5. It was chiefly human activity that brought about the extinction of some 
species and the population decrease of others, directly through unrestricted 
use of firearms and indirectly through the cutting down of forests on the 
inclined piedmont plain. 

Ritual sites similar to those described earlier, containing bone 
material mostly consisting of skulls and horns of east Caucasian and wild 
goats and deer, are also known from Khevsuretia, Tushetia and Dagestan 
(Dinnik, 1914a; Makalatiya, 1935; Maruashvili, 1955). 
2 06 Bones of small mammals in pellets of eagle owls and little owls have 

been collected by the author at two localities in the inner valleys. These 
bones occurred at the bottom of small caverns in limestone rock near the 
207 village of Dzivgis in the Fiagdon gorge. They are not more than 150-200 
years old, completely white and in a good state of preservation. 

The present landscape of this area is composed of turfy dry slopes 
covered with tragacanth, astragali and rocks. Vegetable gardens, hay 
meadows and pastures are located near the river. The height above sea level 
is approximately 1,400 m. The list of the fauna is given in Table 45. 

Approximately 40 fragments of lizards and toads were collected at the 
same locality. 

All the species mentioned live in the present in the dry part of the 
Fiagdon valley. However, it should be mentioned that the occurrence of 
Norway rats, domestic mice and, possibly, migratory hamsters are 
related to the deforestation of the valley. 

Bones of other species were found in the pellets of little owls in the 
ruins of a medieval fortress on the upper reaches of the Doniserdon, the 
right tributary of the Urukh. 

The upper parts of the longitudinal valley are covered with mesophytic 
meadows and birch groves. The altitude is nearly 1,600 m above sea level. 
The fossil species correspond to the fauna of the Recent landscape (Table 46). 



202 



A comparison of Tables 45 and 46 which takes into account the evolution 
of landscapes in the inner valleys indicates that the composition of the mice 
fauna and their distribution in the dry inner valleys changed considerably 
over the last few centuries. 




10 11 12 13 14 15 )6 17 18 

Age in years 



(206: 



FIGURE 95. Relationships between age groups and size of horns of east Caucasian 
goats in Rekom dzuar (length of horns measured along the maximunri curvature) 



TABLE 45. Species and number of bones of mammals from pellets of eagle owls 
and little owls near Dzivgis village 



Species 


Number 
of bones 


Number 
of individuals 




6 

2 

3 

139 

22 

8 

15 

93 


1 


Mustela nivalis 


1 




1 


Cricetulus migratorius 

Microtus arvalis ... 


22 

4 




3 


Apodemus sylvaticus 

Rattus norvegicus 


6 

7 






Total 


288 


45 







203 



The increasing aridity in the valleys, caused by cutting the groves on the 
slopes and grazing of cattle, resulted in the replacement of the mesophilous 
ecological assemblage of rodents (Microtus arvalis, M. major i, 
Chionomys gud) by a xerophilous assemblage (Cricetulus 
migratorius, Mus musculus hortulanus, Lepus europaeus) 
and by the introduction of synanthropic forms (Rattus norvegicus, 
Mus musculus musculus). 




FIGURE 96. Horn covers of goats and deer antlers along the 
walls of Rekom cave 



There are no data on the recent extinction of small mammals in the 
areas under study. Also no remains were found of those ancient species 
of little suslik (Citellus pygmaeus musicus) and Radde's hamster 
(Mesocricetus auratus raddei) which are characteristic of the 
inner valleys of Kabarda in the west and Dagestan in the east. 



208Western Transcaucasia 

In the Holocene human settlements clustered around the terraces of the 
Black Sea coast, mountain river gorges and the shores of water bodies in 
swampy Colchis. 

204 



TABLE 46, Species and number of mammal bones in pellets of little owls from the 
upper Doniserdon 



Species 


Number 
of bones 


Number of 
individuals 




5 
16 

37 

727 

45 
14 


1 


Mustela nivalis 


1 


Microtus gud 

Microtinae 


6 

80 


Including the number identified with 
certainty: 
M.arvalis 


(77) 


M. majori 


(3) 




14 


Apodemussylvaiicus 


8 


Total 


844 


110 



TABLE 47. Species and number of bones of mammals from post-Paleolithic beds of the Akhshtyrskaya cave 



p 

'■ Species 


Number 
of bones 


Number of 
individuals 


Species 


Number 
of bones 


Number of 
individuals 


Domestic 
Dog 


3 
1 
8 
7 
17 
25 


1 
1 

4 
2 
6 

4 


Wild 

Ursus arctos 

Canis lupus 

Meles meles 

Lepus europaeus.... 

Rattus rattus 

Sus scrofa 


4 
4 

3 
1 

27 
5 

25 


2 


Horse 


1 


Pig 


1 


Large cattle 

Goat 

Small . ... 


1 

1 
4 




Cervus elaphus 

Capreolus capreolus 


1 
5 


Subtotal. . . . 


61 


18 




Subtotal 


71 


16 




Total 


132 


34 



The fauna of the Neolithic and younger beds of the Akhshtyrskaya 
cave near Adler are of particular interest. These beds overlie the 
Paleolithic, but it is difficult to subdivide the material from the upper 
layers of the cave (Zamyatnin collections) into the various post-Paleolithic 
2 09 cultures. The bones are light in weight, mostly light yellow in color with 

some dark stains. Some of the fragments have partially lost organic matter; 
in general, they resemble Upper Paleolithic bones. The identifiable material 
consists mostly of epiphyses. The species identified and their number are 
given in Table 47* (Figure 97). 



* The material from the Neolithic and Iron Age layers was combined, because the state of preservation 
of the bones definitely indicates that they were mixed in the layers and during the excavations. 



205 





FIGURE 97. Mammalian bones from post-Paleolithic layers of the Akhshtyrskaya cave 

1 — metacarpal ofUrsus arctos; 2 — radiusofMeles meles; 3 — femur of L ep u s 
europaeus; 4 — femur of Rattus rattus; 5 — metatarsal ofCapreolus capreolus; 
6 — metacarpal ofCapra caucasica 



210 



The composition of food remains at this site is more variegated in wild 
fauna than at other post-Paleolithic localities in the Caucasus, since the 
cave was probably used by hunters and shepherds. 

The absence at the site of mountain ungulates (cave bear, elk, giant 
deer and bison) which were probably driven farther into the mountains is 
quite apparent, as is the appearance of black rat in the Iron Age beds. 

In swampy Colchis the oldest cultural beds containing animal remains 
are known from the excavations of an Eneolithic site. The settlement was 
on a mound, at present submerged in a swamp, near the village of 
Ochamchiri, on the left bank of the Dzhikumur creek (Solov'ev, 1939). 
In the bone collections Gromov has identified a small cow, domestic pig, 
goats and sheep. Wild species include deer, roe deer and fish, particularly 
flounder. It is characteristic that horse is not found among the remains. 

One of the Early Bronze Age settlements which is located near Anaklia, 
close to the mouth of the Ingur River, has been excavated by archaeologists 
of the Zugdinskii Museum. The cultural layer is overlain by Recent silty 
alluvium. According to Kuftin, the very primitive black ceramics place 
this bed in the second millennium B.C. The bones from this site are 
shiny, chestnut in color and slightly fractured. The latter characteristic 
indicates that the ancient settlers of the coast had an adequate supply of 



206 



animals for food. The species identified (Table 48) show that primitive 
animal husbandry and more developed hunting were carried on by the ancient 
settlers. 

TABLE 48. Species and number of bones of mammals from the Bronze Age settlement near Anaklla 



Species 


Number of bones 


Number of individuals 


Domestic 
Pis 


103 
258 

47 


17 


Large cattle 

Goat 


17 

5 


Subtotal 


408 

1 

4 
15 

4 
36 


39 


Wild 
Canis aureus 


1 


Delphinus sp 


3 


Sus scrofa 


3 


Cervus elaphus 


1 


Capreolus capreolus 


4 






Subtotal 


60 


12 






Total 


468 


51 







211 



Individual bones of fish and birds are also known from this locality. 

At present deer do not inhabit the area near the site, but jackals are 
abundant, and boars and roe deer occur occasionally. 

The absence of horses and the small number of dolphin bones, as 
compared with the Gelendzhik locality, is characteristic of the Anaklia site, 

Rich bone material, which has probably been lost, was obtained from 
excavations of the numerous mounds and settlements of ancient Greek and 
later times on the Colchis plains, along the midde and lower sections of 
the Rion. The author has identified remains of a domestic goat, wild boar, 
deer and roe deer, found at the Naokhvamy site, from photographs by the 
archaeologist Kuftin. 

A number of fragmentary antlers of red deer, obtained in the peat areas 
of Colchis during the construction of drainage canals, were deposited in 
the museum of the town of Poti. 

A very important deposit of food remains was exposed in the excavations 
of Holocene (Neolithic and latest) beds at the Sagvardzhile site, on the 
eastern margin of the Colchis plain (Kiladze, 1953). 

The collections from these beds which have been studied by the author 
contain the following wild species in order of decreasing abundance: * 



Sus scrofa 
Cervus elaphus 
Capreolus capreolus 
Bison bonasus 
Capra caucasica 
Rupicapra rupicapra 



Ursus arctos 
Equus caballus 
Felis silvesrris 
Panthera pardus 
Castor fiber 



Judging by the state of preservation, this material, which is under study by Burchak-Abramovich, 
undoubtedly also includes Paleolithic bones. 



207 



The occurrence of the beaver bones confirms that it was in Colchis that 
this species survived the Pleistocene in the Caucasus. 

Of the many western Transcaucasian monuments of ancient architecture 
and sculpture ornamented with animal design, the Bagrat temple (1003 A.D. 
in Kutaisi is particularly noteworthy. A scene showing cheetahs and 
panthers attacking a horse or a kalan and a goat is depicted on the heads of 
the temple columns. Taking into account the broad base of medieval 
Georgian culture it is obviously difficult to maintain on the basis of these 
ornaments that kulans and wild horses existed in Imeretia in the 10th 
century (Figure 98). 



■"' ■ , ,'<l!«*1^ ^**** \ 







FIGURE 98, Scene on the head of Bagrat temple column — panthers attacking a goat 

Photograph by author, 1956 



212 



The latest developmental stages of the ecological assemblages of small 
mammals in Imeretia can be traced through bones brought by eagle owls 
into small caverns in the walls of the canyons which cut through the 
Imeretian Plateau. In 1952 we collected material (identified in Table 49) 
from close by the Sakazhia cave in the Tskhali-Tsiteli gorge which 
shows the interesting, simultaneous occurrence of jackal, hare, Norway 
rat and fat dormouse, species not known from the Paleolithic beds of 
Imeretian caves. 

As opposed to eastern Caucasia, no collections of game animal skulls 
have been preserved from the mountainous section of western Transcaucasia 
in the Ingur, Rion and Tskhenis-Tskhali gorges which were inhabited by the 
Svans. 

Radde, it is true, wrote in 1866 that in 1864 he had observed an 
accumulation of horns of Caucasian goats, piled by hunters near the chapel 
of the village of Chibiana (Ushkul'), in the upper reaches of the Ingur. 
He estimated the number of horns at "several thousands. " In 1948 I could 



208 



find only 7 pairs of horns and horn stems of goats (6 of С a p r a 
caucasica and 1 ofC.cylindricornis)in the altar of the chapel in 
the village of Zhabeshi and in the altar of the Lamardiya monastery above 
Ushkul'. No confirmation of the past existence of "thousands of goat horns" 
could be obtained from old residents or from the local authorities. But it 
seemed that the hunters of this village used to store horns and blade bones 
of Caucasian goats and chamois under the roofs of their observation towers, 
By 1948 the collections were mostly abandoned or given away and no new 
material was introduced due to the extermination of the aninaals. 



TABLE 49. Species and number of Holocene bones of mammals from pellets of 
eagle owls in the Tskhali-Tsiteli gorge 



Species 


Number of 
bones 


Number of 
individuals 


Erinaceus europaeus 

Canis aureus 


37 
1 
2 

12 

74 


4 
1 


Lepus europaeus 


1 


Glis glis 


3 


Rattus norvegicus 


12 


Total 


126 


21 







It was the practice of the Svans, as opposed to that of the Ossetians, 
to cut the goat horns off with only a small piece of frontal bone. Even now 
goat horns and bear paws can be seen attached to the porches of houses 
and embedded in the stone walls of houses made of slate. 

Thus, the studies of Holocene bones in western Trancaucasia indicate 
an abundance in the Neolithic and later epochs of larger game animals: 
boar, deer, roe deer, bison and bear, as well as the European beaver found 
in Colchis. The appearance of new species, unknown in the Pleistocene 
(Paleolithic) beds (jackal, Norway rat and black rat) is highly 
characteristic. The composition of the donaestic fauna and the nature of 
their deposition indicate that, in general, the landscape of Colchis and the Black 
Sea coast was very similar to the landscape of the beginning of the present 
century. 



Central Transcaucasia 

Reconnaissance excavations were carried out in 1952 by Lyubin above 
the overhangs of the natural landmarks of Saroty-Or and Sharshiat-Kadzakh 
near Staliniri. We have identified in Lyubin's collections bone fragments 
of small domestic pigs, bulls, goats and sheep, and teeth of European brown 
bears and foxes. Rich collections were also obtained in 1955-1957 from the 
Holocene beds of the Kudaro caves on the upper Rion. 

In the valleys of the Kura and the Aragva rich bone material was 
excavatedby archaeologists of the Caucasian Museum from multilayered 



209 



213 cemeteries and towns at Samtavro and Armazi, near Mtskheti.* A comparison 
between photographs of this material in the archives and exhibit material 
of the Upper Bronze Age (9th-8th centuries B.C. ) show that these were 
bones of horses, cows, sheep, goats and boars, and amulets made of canines of 
European brown bear and incisors of beavers. Skeletons of martens (M ar t e s 
f о i n a) and steppe voles (Microtus socialis), which died at a later 
period, have recently been discovered. Objects made of red deer antlers 
are very common, e.g., the so-called "spear rectifiers. " Elegant bronze 
plates and trinkets of this period from Samtavro and Armazi often depicted 
dogs, foxes, lions, panthers, hares, deer, goitered gazelle, rams, 
Caucasian goats and Bos primigenius. 

Taking into account the paleontological data for Transcaucasia as a whole, 
these artistic representations confirm the existence of some large game 
animals in the Caucasus during the first millennium B.C. These animals 
became extinct only in the Middle Ages. 



Eastern Transcaucasia 

On the Kura Plain and in the foothills of the Greater Caucasus, Karabakh 
and Talysh, Holocene deposits of bones occur in diluvium, alluvium, caves, 
asphalt crusts and cultural layers of ancient settlements. 

There is a record of a horn of a red deer from the diluvium of a mountain 
slope near the village of Alty-Agach, near the eastern margin of the 
Bolshoi Range (Deminskii, 1901). Many occurrences are concentrated on 
the Apsheron Peninsula. 

Kovalevskii found in 1941 a skeleton of a striped hyena (Hyaena 
hyaena) at a depth of 4 m in Holocene loess loams in the lower section 
of the Sumgait rivulet, north of Baku. The skeleton was buried under a 
cultural layer containing medieval pottery and under a bone-bearing bed of 
an older settlement, overlain by the sediments of the Recent Caspian. From 
the stratigraphic position of the skeleton and its state of preservation 
it can be dated as Lower Holocene. 

A skull of a very old striped hyena, bones of a horse and of a Caspian 
seal (Phoca caspica) were found by Kovalevskii in the Bailovskaya 
cave on the southern outskirts of Baku (Bogachev, 1938b). 

We found two leg bones of goitered gazelle (Gazella subgutturosa) 
in blocks of asphalt-impregnated surface loams overlying the Middle 
Quaternary beds at Binagady. The bones are fresher and less impregnated 
with oil than the Pleistocene bones. In the asphalt crusts of Mount Kir- 
Kishlag we discovered bones of hare and wolf. Skeletons of small rodents, 
jerboas (Allactaga e 1 at e r), hamsters (Cricetulus migratorius) 
and voles (Microtus socialis), which died in asphalt, also occur in 
this locality and at Mount Zigil'-Pirya (Vereshchagin, 1951b). 

The numerous Holocene settlements and burials of Azerbaidzhan (Passek 
and Latynin, 1926) are responsible for the rich bone collections from the 
cultural layers. 

Gummel' (1948), who excavated the Bronze Age mounds on the sloping 
plain west of Khanlar (formerly Helenendorf), mentions numerous bones of 

' This material remained unidentified and was subsequently lost. 



210 



Bos, sheep, goat and domestic pig. The number of horse bones was 
negligible. 

Excavations of the settlements and the burial near ancient 
Mingechaur on the right bank of the Kura, near a hydroelectric 
power plant, revealed cultural layers dating from the second millennium 
B.C. to the early Middle Ages. The very rich bone material from this 
locality has not yet been completely identified. Our preliminary study of 
these materials indicated the presence of numerous bones of horses, asses, 
cows, sheep, goats and, more rarely, domestic pigs. The wild animals 
are represented by red deer (Cervus elaphus), goitered gazelle 
(Gazella subgutturosa) and boar (Sus scrofa), all of which were 
until recently quite common in the region. 

We found a particularly interesting burial near Mingechaur with bones 
of black rat (Rattus r at t u s) and horse, ass, large cattle and 
boar buried in jugs. This type of burial, which included sacrificed domestic 
goats and grain, is usually dated as middle of the last millennium B.C. In 
view of the fact that the burial field has been covered for many centuries 
by a semidesert saltwort vegetation, the rats could only have inhabited the 
field during a time when it was settled and irrigated or at the time when 
the cult of mass burial flourished. Their occurrence verifies assumptions 
of the early appearance of black rat in Transcaucasia (Vereshchagin, 1949c). 

TABLE 50. Species and number of mammal bones from ancient settlement near Sumgait 



Species 


Number of bones 


Number of individuals 


Domestic 
Horse 


6 
11 

12 


1 


Large cattle 


3 


Sheep 


3 


Subtotal 


29 

2 

3 


7 


Wild 
Lepus europaeus 


1 


Gazella subgutturosa 


1 






Subtotal 


5 


2 






T otal 


34 


9 







A very old settlement was discovered by Kovalevskii in 1941 in Holocene 
loams near the mouth of the Sumgait rivulet. Food remains were 
found at a depth of 3 m under the cultural layer containing medieval pottery; 
they were well preserved under the very dry burial conditions. The species 
found are given in Table 50. 
215 At the same locality ten fragments of bones of mute swan (C у gnus 

о lor L. ) and goose (Anser anser L. : three specimens) were found. 

According to Kovalevskii, the bone -bearing bed is overlain by sediments 
of the present-day Caspian and therefore antedates the last Holocene 
transgression of the Caspian, i.e. , early first millennium B.C. 



211 



Large collections of bones were made in 1953-1954 by archaeologists of 
the Azerbaidzhan Academy at the excavations of the fortress Uren'-Kala 
(Bailakan) in the center of the Mil'skaya steppe. We observed the layers 
from the Middle Ages at this locality to be characterized by sheep, large 
cattle, goat and horse, with a small admixture of goitered gazelle (Gaze 11a 
subgutturosa). 



TABLE 51. Species and number of mammal bones from the medieval settlement 
on the Baku Fortress hill 



Species 


Number of bones 


Number of 
individuals 


Domestic 

Dog 

Cat 


3 

1 

39 

15 

3 

111 

202 

3 

223 


3 
1 


Horse 

Ass 

Camel 


10 
5 
2 


Large cattle 

Sheep 

Goat 


14 
29 

2 


Small cattle (no closer identification) 


15 


Subtotal 


600 

1 

1 

1 

48 


81 


Wild 
Phoca caspica 


1 


Vulpes vulpes 


1 


Equus hemionus 


1 


Gazella subgutturosa 


18 


Subtotal 


51 


21 






Total 


651 


102 







216 



Bone material from beds of the Middle Ages was also obtained by 
Leviatov in excavations of the Shirvan-Shah Palace courtyard in Baku. 
The excavations revealed that small settlements successively occupied 
the hill of the contemporary Baku Fortress. The beds which have been 
studied are dated by the pottery as 9-13th century A. D. This is the time 
of the end of the Arabic caliphate, the age of Nisami and the beginning of 
the Mongolian invasions into eastern Transcaucasia. Part of the bone 
material buried at the site probably resulted from epidemic death among 
the cattle, since complete scapula and forelimb bones are common. 
However, most of the bone fragments are food remains which were well- 
preserved in the dry ground. 

Bones of domestic animals — sheep, cows, horses and asses — 
characteristically predominate. Bones of wild species — goitered gazelle, 
kulan, fox and seal — are not numerous (Table 51). 



212 




FIGURE 99. Ruins of Chukhur-Kabala fortress 



Photograph by author, 1951 



From the same beds the following birds have been identified: С у gnus 
sp. , Anser sp. , Otis tarda L. , O. tetrax L. , domestic fowl, 
totalling 21 fragments of 10 individuals. In addition bone identification was 
made of the following fishes: Acipenseridae (25 fragments of 4 individuals) 
and Cyprinidae (58 fragments of 8 individuals). 

Most of the goitered gazelle bones occur in the 10 — 11th-century beds. 
The metatarsus of a kulan was found in the 13th-century bed. 

The findings in ancient Baku confirm an abundance of goitered gazelle 
and kulans in eastern Transcaucasia. The hunting of seals and aquatic 
birds, and fishing in the Baku bay have also been established. 

Bones and complete skulls of rats (Rattus norvegicus) are common 
in recent layers, dated 15-16th century A. D. However, the contemporaneity 
of the bones and the beds is not established, since rats may often penetrate 
the deep layers by descending through crevices and holes in the ground. 

We studied bone material from the beds of the ancient Chukhur- 
Kabala fortress (Figure 99), located on the northern margins of the 
Kura Plain in the Alazan-Agrichai valley and destroyed by the Mongolians 
in the 14th century (Krymskii, 1934). The ruins of the fortress and the 
towns are located between two left tributaries of the Turyan-Chai in thick 
broadleaf forest. During the period of the maximum development of these 
towns the trees were probably felled, since fragments of square bricks 
217 and pottery occur in the vicinity, particularly in the beds of the Kara-Su 
springs which flow through the alder and wing nut thickets. 



213 



Bones of wild and domestic animals, mixed with pottery, bricks, ash 
and earth, fill deep wells in the fortress, which are at present being activex^ 
eroded from the east by the Kara-Chai rivulet. The identified bones of 
domestic and wild animals which we collected in the talus of the cliff and 
in the rivulet bed are given in Table 52. 



TABLE 52. Species and number of mammal bones from the Chukhur-Kabala 
fortress beds (Middle Ages) 



Species 


Number of bones 


Number of 
individuals 


Domestic 
Ass 


1 
3 

13 
1 

38 


1 


Horse 


2 




2 


Buffalo 

Small cattle 


1 
5 


Subtotal 

Wild 
Cervus elaphus 


56 

1 
1 


11 

1 


Capreolus capreolus 


1 


Subtotal 


2 


2 






Total 


58 


13 







Special excavations would undoubtedly increase the number of wild 
species specimens. In 1946-1947 we collected bones of small animals on 
the northern margins of the Kura Plain and in Apsheron from owl pellets. 
The material was collected between A Id zhigan - С hai and Geok-Chai 
in short-lived caves in the cliffs of the Tertiary hills which separate the Alazan- 
Agrichai valley from the Kura Plain. Study of the fossil material showed 
that it was identical with the modern assemblage of insectivores and rodents 
known from samples obtained with traps (Table 53). 

We obtained more complete materials in a study of past and present 
distribution of small mammals on the Apsheron Peninsula, which enabled 
us to compare the composition of the present and the Pleistocene faunas. 

A narrow crevice (Figure 100) which had been occupied for several 
centuries by little owls was discovered by Burchak-Abramovich in 1944 in 
the limestone rock east of Cape Bil'gya on the northern coast of the 
Apsheron. Bones of insectivores, rodents, birds, reptiles and amphibians, 
and chitin of beetles from decayed pellets formed a layer 10-12 cm thick 
218 in the cavern. The bones in the upper layer were brown in color; those 
in the lower layers were gray-brown. Groniov, in 1944, and I, in 1945, 
collected up to 5,500 bones of mammals in the crevices. There were fewer 
bones of birds, lizards and toads. 



214 



TABLE 53. Species and number of bones of small mammals in owl pellets from Geok-Chai 



Species 


Number of bones 


Number of individuals 


Hemiechinus auritus 

Crocidura russula 


2 

33 

4 

5 

106 

16 

36 

15 


1 
7 




1 


Cricetulus migratorius 

Microtus socialis 


2 
19 




6 


Apodemus sylvaticus 

Meriones erythrourus 


9 
1 


Total 


217 


46 




FIGURE 100. Resting place of little owls in the rocks of Cape 
Bil'gya 



As old and plentiful as the accumulations found in owl prey are, they 
contain an essentially modern faunal assemblage, very similar to that which 
occurs at present on the northern Apsheron and is known from extensive 
samplings obtained with traps (Table 54). 

In both the owl-pellet accumulation and the trap samplings, there is a 
disproportionate representation of certain rodent species, which reflects 



215 



in the one case the selectivity of the owls and in the other the specific 
instructions under which the personnel of the anti-plague station operated. 

There is no doubt that the ecological assemblage of smaller mammals 
of the Apsheron remained relatively stable over the last part of the 
Holocene. The study of the distribution of Recent mammals showed that 
219 the European hedgehog, forest dormouse, field mouse and steppe vole are 
relict species which have survived in a few sections of the peninsula from 
a more humid climatic stage, as have some other vertebrates, in particular 
Rana ridibunda L. and the pond turtle Emys orbicularis L. 
(Vereshchagin, 1949c) (Figure lOl). 



TABLE 54. Species and number of smaller mammals on Apsheron 



Species 


Caught by owls in the 

last three to four 

centuries 


Trapped in 1939-1940 


Erinaceus europaeus 


8 
2 

147 

83 

63 

1 

33 

20 


Rare 


Hemiechinus auritus 


Common 


Crocidura russula 


Common 


Pachyura etrusca 


Not recorded 


Dyromys nitedula 


Rare 


Cricetulus migratorius 


73 


Microtus socialis 


115 


Meriones erythrourus 


2,123 
8 


Allactaga elater 


A. williamsi 


50 


Mus musculus 


81 


Apodemus sylvaticus 


6 


Rattus norvegicus 


2 






Total 


357* 


2,458 







' The number of individuals is estimated from the number of lower jaws. 



The graphic art of nomadic tribes provides indirect evidence of the 
occurrence of large mammals in the historical epoch in eastern Transcaucasia. 

A number of ancient campsites of nomads occur under the cliffs of 
Apsheron limestone in Kabristan, southwest of Baku. Ancient craftsmen 
have left line drawings of lions, red deer, wild and domestic goats, 
domestic bulls and horses on the picturesque rocks and overhangs of Mount 
Kyzyl-Kum, Beyuk-Dag and Dzhangir-Dag, which were probably also 
religious sites (Vereshchagin and Burchak-Abramovich, 1948). Some of 
the drawings are probably Neolithic, some Bronze Age and some belong 
to the present era (Figure 102). 

The studies of Holocene burials in eastern Transcaucasia may be 
summarized as follows. In the postglacial epoch the following species 
appeared in eastern Transcaucasia: wild forms — striped hyena and 
possibly lion; rodents — black and Norway rats; hoofed mammals' — kulan 
and goitered gazelle. A number of large mammals, characteristic of the 
Middle and Upper Pleistocene, disappeared from the area probably at a 
somewhat earlier stage. 



216 



49°00' 



4Э°30' 



SCOO' 

— т 



PRESENT DISTRIBUTION OF SOME 

RELICT SPECIES ON THE 

APSHERON PENINSULA 

AND IN KABRISTAN 




49°00 



«"30' 

FIGURE 101 



50*00' 



(220), 




FIGURE 102. Drawing of goats and other animals on rocks of Dzhingir-Dag 

217 



Until recently some ungulates (e, g. , deer and goitered gazelle) 
were much more abundant; the deer inhabited the tugai forests of the middle 
reaches of the Kura River. 
221 The occurrence of relict species (European hedgehog, fat dormouse, 

field mouse, Rana ridibunda and pond turtle) on the Apsheron 
indicates that the climatic regime remained relatively stable during the 
historical epoch. Their distribution also reflects changes in the landscape 
caused by man. 



Lesser Caucasus Upland 

Postglacial mammals of the Armenian Plateau occur in diluvial 
sediments, tuffaceous conglomerates, lake sediments, caves, cultural 
layers of settlements and burial grounds. 

Localities in diluvium on the shores of Lake Sevan. 
As early as 1926 Shelkovnikov brought to the ZIN rounded fragments of a 
bison skull (Bison bonasus subsp. ), collected from postglacial, 
tuffaceous conglomerates at the Zanga source in Lake Sevan (Gromova, 
1935c). The state of preservation of this material shows that it may be 
Bronze Age . 

A large deer skull with excellently preserved antlers was collected from 
the diluvium of a slope near the village of Kabakhlu in the Azizbekov area 
in Armenia (Dal'. 1947a). 

A fragment of the skull of a suslik (Citellus cf. xanthoprymnus) 
was collected from the alluvial loams on the right bank of the Zanga, 
6.5-7.0 m below the surface, in the vicinity of Arzni. The loams overlie 
a lava flow, presumably of the third to second millennium B.C. The time 
of the suslik is considered to be somewhat more recent (Dal', 1949a). The 
present eastern boundary of the suslik range is 40 km to the west. 

Many animal bones were deposited in the littoral zone of Lake Sevan, some 
from animals which fell through the ice and drowned, some brought by the 
mountain streams. A large proportion of the bones are food remains of 
ancient hunters and cattle herdsmen who lived in pile dwellings along the 
lake shores. From the vicinity of such structural remains at the Zanga 
source two pairs of large antlers of red deer were collected and 
transferred to the Caucasian Museum (Bayern, 1871). Pile structures were 
also found near Novo-Bayazet and near Lake Gilli (Lalayan, 1929). 

In 1947-1948, when the level of the lake started to fall. Dal' (1950b) 
collected numerous bone fragments on the exposed part of the northern 
shelf of the lake. The material included bones of deer, European wildcat, 
boar and bear. A large collection of fragments of antlers and skull fragments 
of boars and turs was obtained from a 1.5-m depth in the shallow-water 
bay in the Artanishskii Gulf. 

Material from the gray river sands of the Sevan coast (the lower reaches 
of the Gavaraget River, near the Agbulakh rivulet and Cape Sary-Kai) 
appeared to be older; it included fragments of antlers, fragments of skulls 
and vertebrae of large bulls, similar in size to Bos primigenius and 
bison. A fragment of skull of a dwarf tur (Bos cf. mi nut us) was 
collected from the same sediments near Cape Sary-Kai (Dal', 1950a). 



218 



At our request Khaveson collected some bones on the Lake Sevan shores, 
mostly on the northwestern shore near Shordzha, and on the southern shore 
near Cape Noraduz and Sary-Kai. 
222 We studied both collections. Most of the fragments in Dai's collection 
are brown and light brown in color; in Khaveson's collection the material 
is mostly of a light color. The bones seem heavy and permineralized, 
which, however, is only due to the limey deposit on the surface. Bones 
from the surf zone are strongly rounded and beige in color; those from the 
sediments in quiet water and from silt are complete and of a dark color. 
The deer remains mainly consist of antlers, often in a state of decomposition 
due to their exposure to water and subsequent weathering on the shore. 
Complete metapodia and jaws of domestic bulls and buffalos have been found. 
From the degree of coloration and loss of organic matter the material can 
be subdivided into two age-groups. The species identified and their 
proportions are given in Table 55. 



TABLE 55. Species and number of bones of mammals from Lake Sevan shores 



Species 


Semifossil 


Recent 


Subtotal 


Domestic 

Dog 

Cat 


12 
1 

61 
6 
3 

70 
5 

13 


8 

23 
2 
3 

67 
2 

14 


20 
1 


Horse 

Pig 

Camel 

Large cattle 

Buffalo 

Small cattle (no closer identification) . . . 


84 

8 

6 

137 

7 

27 


Subtotal 


171 

1 
1 

5 

19 


119 

1 

1 

33 


290 


Wild 
Vulpes vulpes 


2 


Ursus arctos 


1 


Meles meles 


1 


Sus scrofa 


5 


Cervus elaphus 


52 






Subtotal 


26 


35 


61 






Total 


197 


154 


351 







The occurrence of dromedary is particularly interesting. The state of 
preservation of the forearm, scapula and mandible (Figure 103), which 
are colored light brown to a depth of 4-5 mm, indicates that the bones are 
quite old (3,500-4,000 years). 

Khaveson (l954b), on the basis of teeth indexes, identified the jaw 
of the camel as belonging to a wild form — Camelus dromedarius 
dahli. Confirmation of this identification based on more complete material 



219 



would indicate an extensive developnnent of xeromorphic landscapes of the 
Armenian Highland at some early stage of the Holocene. 
223 It seems very doubtful that wild camels could migrate into a mesophytic 
landscape of the present upland type in the vicinity of Lake Sevan. Other 
wild forms (bear, boar and deer) are more characteristic of the mesophytic, 
i.e. , meadow and forest, environment than of the upland steppe and desert. 

Material from burials and settlements. It has been possible 
to date more accurately the bones from burial grounds and ancient towns 
on the Lake Sevan shores, the Karabakh plateau, the Zanga shores, Debed- 
Chai ravine and the upper reaches of the Khram, near Tsalka on the 
Trialet ridge. 

According to Piotrovskii (1949, p. 92), the oldest Holocene site is the 
Eneolithic, on the bank of the Zanga River near Shengavit, which 
was excavated by Baiburtyan in 1936-1938. The site was dated as second half of 
the third millennium B.C. Dal' has identified in the site material bones of wild 
goats, horses and fish; the horses were probably wild varieties. 

Remarkable material, including antlers of red deer, was collected by 
Kuftin (1939) in 1930 from Eneolithic (third millennium B.C. ) mound burials 
on the Trialet ridge in the upper reaches of the Khram. 

A lower molar of a bear (Ursus arctos m e r i d i о n a 1 i s) was 
collected by Kuftin from an Upper Eneolithic single burial near the village 
of Ozny in the Tsalka area. 




FIGURE 103. Lower jaw of С a melus cf. d ro m e d a r iu s from 
the shores of Lake Sevan 



More recent burials from the middle of the second millennium B.C. 
contained fragments of skulls, metapodia and phalanges of giant bulls, 
nearly the size of В о s primi genius, as well as bones of dogs, horses, 
boars, sheep and goats. All the bones are stained with humus; they are 
dark brown in color, and partially decayed. Kuftin thought that venerated 
human remains were carried on bull-drawn carts to the grave site. The 
bulls were then slaughtered and eaten, and the skins, skulls and limb bones 
were buried with the corpse. These mounds yielded a silver bucket 
depicting red deer and chamois and a gold goblet depicting lions. 



1704 220 



The same type of burial with remains of a bull was found in 1948 in 
Kirovakan (Piotrovskii, 1950). The skull, better preserved than the 
one at the Trialet locality, indicates that the primitive tribes who 
built burial sites of this type had begun the domestication of В о s 
primigenius. A pair of roaring lions, similar in style to Hittite 
drawings, are depicted on a gold cup found in the grave. 
224 It is known from Resler's (1896) descriptions "of the sites of huge fires" 
that "great accumulations of ashes and burnt animal bones —horses, rams, 
dogs, boars, buffalos, and bulls — " occur in the mounds and burial sites at 
Archadzor near the village of Khodzhala on the Karabakh Plateau. The 
presently accepted date of these monuments, which are related to the 
Assyrian culture, is early first millennium B.C. (Piotrovskii, 1949). 
Skulls of horses, bones of rams, camels and wild animals were found in 
the dolmens excavated by Belek and Ivanovskii in Karabakh and near Kedabek 
(Sysoev, 1925). 




FIGURE 104. 

Skulls: 1 — Vulpes vulpes; 2 — Martes foina; 3— Lepus europaeus 
from Urartu burials on the shores of Lake Sevan 



221 



The excavations of the Urartu fortress Tsovinar on the southern shore 
of Lake Sevan (Piotrovskii, 1944) revealed bones of domestic pigs, domestic 
bulls and sheep. Large Urartu burial fields on the eastern shore of Lake 
Sevan near the village of Zagalu were excavated in 1905-1908 by Lalayan 
(1929). Big earth mounds of three types, which cover stone mausoleums 
used for common burials, were found there and near the villages of Nor- 
Bayazet and Adiaman. By the transitional Bronze-to-Iron-Age material 
found there, these burials have been dated as early first millennium B.C. 
225 (Piotrovskii, 1949, p. 86). The sites contained skulls of martens (M a r t e s 
latifrons= Martes foina) and badgers (Meles urartuorum = 
Meles meles minor) (Satunin, 1907b). From the same sites, 
Shelkovnikov in 1930 collected a mandible of a horse, a skull of a young 
pig, two skulls of foxes (Vulpes vulpes), four skulls of martens 
(Martes foina) and a skull of a hare (Lepus europaeus), all of 
which he donated to the ZIN. 

The burial of the bodies of wild and domestic animals was of ritualistic 
importance to the Urartus. Lalayan mentions that he invariably found 
skeletons of horses, large cattle, sheep and goats in Urartu graves which 
he disposed of at the site. It is not known whether complete bodies of wild 
animals or only their skulls were buried (Figure 104). 

TABLE 56. Species and number of mammal bones from Urartu fortress Teishebaini 



Species 


Number 

of 

individuals 


Species 


Number 

of 

individuals 


Domestic 


3 
8 
6 
6 

22 
2 

25 
7 


Martes foina 


1 




Felis lybica 


1 


Doe 


Cricetulus migratorius. , . . 
Mus musculus 




Horse 


2 
2 


Ass 


Apodemus sylvaticus 

Meriones sp 

Capra aegagrus 




Pig 


4 


Large cattle 


1 




1 


Buffalo 


Ovis gmelini 


1 


Sheep 

Goat 


Gazella subgutturosa 

Cervus elaphus 


2 




4 




79 

4 
1 






Subtotal 


Subtotal 






24 








Wild 


Total 


103 


Crocidura russula 






Hemiechinus auritus 





Ornaments from the graves usually depict red deer. A running bear 
is engraved on a sard seal collected in one of the graves. 

Excavations of the Urartu fortress Karmir-Blur (Teishebaini) 
on the Zanga River near Yerevan (Piotrovskii, 1944, 1949) produced bone 
material which reflects some features of the animal husbandry and the 
faunal peculiarities of the Armenian Highland in the middle of the first 
nnillennium B.C. 



222 



Teishebaini, an important administrative center of the Urartus, was 
conquered in the 7th century B.C. by the Scythians who invaded 
Transcaucasia. Domestic and wild animals and various objects were buried 
under falling roofs and walls as the fortress burned. Dal' (1952), who has 
studied 4,667 bones from the Teishebaini excavations, listed the species 
given in Table 56. 

From his knowledge of the conditions of the burial and his personal 
observations at the excavation site, Dal' concludes that gerbil and long- 
eared hedgehog migrated to the ruins of the fortress at a later time. The 
stone marten (Martes f о i n a) and the spotted cat (Felis lybica) 
might have been domesticated species which were used to control rodents. 
226 White -toothed shrew, hamster, field and house mice belong to the local 

synanthropic forms which at that time invaded human dwellings. Bones and 
objects made of horns of wild goat, ram, deer and goitered gazelle give 
some idea of the hunting traditions in the northern part of Urartu. This 
incidence of goitered gazelle bones is the first documented evidence of the 
occurrence of this gazelle in the middle section of the Araks valley. 

Material from shelters of predators. New paleontological 
materials of the postglacial epoch were found by zoologists during studies 
of accumulated food remains of predators in the caves and smaller caverns 
of the Lesser Caucasus upland. 

Polyakov (1882), in his survey of the caves along the shores of Lake 
Sevan in 1879, found a skull of a weasel, jaws of voles, mice and ducks in 
a cave at Mount Peschanaya Gor. 

In his study of the present vertebrate fauna of the Saraibulakh (Urtskii) 
ridge, Dal' (1940b, 1954a) collected and identified the remains of animals 
accumulated by preying eagle owls and wolves in four karst caves. (Some of 
the bones belonged to animals which entered the cave and died there. ) The 
richest material was collected in a cave in the third gorge from the west, 
in the western part of the ridge, above the village of Shagaplu. The entrance 
into the cave is on the right wall of the gorge, 11m above the ravine floor 
(1,750 m above sea level). The combined length of the sections of the cave 
which were studied is 41.5 m. The animal bones were scattered and were 
mainly found in depressions and chambers, occurring in large numbers 
in the clayey surface layer. In this cave, 1,850 bones, belonging to at least 
660 individual reptiles, birds and mammals, have been collected. 

The state of preservation and the age of the bones vary. Some bones of 
domestic animals which remained on the surface were better preserved 
than those which were embedded in the ground. The bones are grayish 
yellow in color. Calcareous crusts up to 1.3 mm thick and stalagmites up 
to 5 mm high formed on some of the bones. The oldest bones dissolve 
completely in hydrochloric acid. Dal' estimates their age from 2,000- 
4,000 years. They include bones of jackal, domestic dog, fox, stone 
marten, tiger polecat, badger, hare, ass, horse, deer, bull, wild goat 
and Armenian mouflon. 

We have studied the skulls of badgers from this cave (Vereshchagin, 
1951b) and found them to be closer to that of the Urartu form of badger than 
those from the Sevan burials. This fact makes it seem possible that some 
of the bones are older than the age assigned to them by Dal'. The jaws 
of a large pika, found in the far corner of the cave, are among the oldest 
bones. The remains of white-toothed shrews, Myotis, foxes, weasels. 



223 



martens and rodents are of modern types. Most of them were washed out 
of the pellets of eagle owls by water. 

The identification of the material from the four caves resulted in the lon| 
list of species given in Table 57. Bones of larks, chough, pigeons, 
Hungarian and rock partridges, Ophisaurus apus and lake frogs were 
also found. 

The location of the caves in a transitional landscape zone and the ways 
by which the animal remains were introduced into them preclude their 
usefulness as a basis for any far-reaching conclusions on changes in the 
climate and landscapes during the Holocene. 

The list of wild species of the Saraibulakh ridge shows that large species 
(jackal, bear, pika, kulan and deer) do not occur there in the present; 
227 their disappearance can undoubtedly be attributed to man. 



TABLE 57. Species and number of mammal bones from caves on the Saraibulakh (Urtskii) ridge 



Species 


Number 
of bones 


Species 


Number 
of bones 


Domestic 
Dog 


11 

8 

11 

15 

28 
26 


K4eles meles 


14 


Felislybica 


1 


Ass . . 


Lepus europaeus 


98 


Large cattle 


О cho tona sp 

Allactaga williamsi 

Meriones persicus 

Cricetulus migratorius .... 
Mesocricetus auratus 


3 


Sheep 

Goat 


9 
135 




135 


Subtotal 


99 

1 

1 

3 

3 

4 

35 

16 

3 

15 

1 




794 
140 


Wild 




6 


Erinaceus europaeus 

Crocidura russula 


Arvicola terrestris 

Ellobius lutescens 

Equus hemionus 


2 
323 




4 




Cervus elaphus 


24 




Capra aeeaerus 


85 




Ovis gmelini 


76 










Subtotal 


1,931 


Martes foina 




Vormela peregusna 


Total 


2,030 







Taken as a whole, the faunal complex of the region, even in the oldest 
times, was very close to the present faunal complex, which is characteristic 
of the arid plateau landscape of southwest Asia. 

In southeastern and central Armenia, Sosnikhina (1947, 1948) did a 
thorough study in 1946-1949 of bones from eagle owls' pellets in three 
different altitudinal vegetation zones and landscapes. These were: l) zone 
of spiny tragacanth formations (1,500-1,600 m above sea level); 2) zone 
of upland fescue steppe and arid thin forests (2,100-2,200 m above sea level) 
in the Azizbekov area and near Lake Sevan; 3) zone of mountain meadows 



224 



(2,600 m above sea level) in Nakhichevan A.S.S.R. The material, consisting 
of several thousand bones, illustrates very well the distribution of small 
mammals on the high plateaus and the effects which changes in climate 
and landscape can have on formation of ecological assemblages. These 
data are summarized in Table 58. 

Most of the bones studied probably accumulated during the last centuries, 
possibly only during the last decades. 

The two main assemblages of small mammals, the xerophilous and the 
mesophilous, either replace or supplement each other, reflecting in part 
the landscape of the region and in part the distribution and hunting behavior 
of eagle owls. 

These data also indicate some etiological features of the behavior of 
eagle owls: the length of time they occupy their shelters, their habits of 
preying on a wide variety of species, choosing their prey in various 
biotopes, and leaving their pellets near the hunting ground. In the winter some 
of the eagle owls probably migrate from the highlands to the valleys, and , 
228 in the absence of shelters on the hunting grounds, they may migrate to 
another area to find day resting places. 



TABLE 58. Species and relative abundance (in%)of mammalian bones in pellets of eagle owls of 
Armenian Plateau 



Species 



Chirakhan 
ravine (1,500- 
1,600 m above 

sea level) 



Southern slope 
of Sevan ridge 
(2,000-2,100 m 
above sea level) 



Dzhagri ravine 

(2,200 m above 

sea level) 



Slof)es around 

Lake Aknalich 

(2,600m above 

sea level) 



Erinaceus europaeus 

Crocidura russula 

C. leucodon 

Sorex minutus 

Neomys fodiens 

Myotis oxygnathus 

Mustela nivalis caucasica ... 

Lepus europaeus 

Ochotona sp 

Allactaga williamsi 

Dyromys nitedula 

Apodemus sylvaticus 

Meriones sp 

Cricetulus migratorius 

Mesocricetus auratus 

Microtus arvalis 

M.socialis 

M . m a j о r i 

M. nivalis 

Arvicola terrestris 

Ellobius lutescens 



4.5 
4.5 



31.9 

9.1 

18.2 

9.1 



4.5 
18.2 



2.5 



2.5 

2.5 
82.5 
10.0 



0.7 
1.5 
1.1 
0.3 
0.3 
0.3 
0.3 
0.7 

1.2 
0.3 
5.0 

9.6 
22.5 

44.2 

4.9 
4.9 
0.7 
2.0 



2.4 



1,2 

4.7 



17.3 
62.5 

2.4 
2.4 
4.7 
2.4 



225 



The occurrence of pika bones is of interest, sinces this species has 
probably become extinct in modern times on the Armenian Plateau. 

In the upland semidesert of the Araks valley, bones from the pellets of 
owls and eagle owls were collected by Sosnikhina (1950b) in a cave of the 
volcanic Mount Karakhan (900 m above sea level), near the village of 
Novyi Shakhvarut, west of Yerevan. In 1947 we collected material in caves 
and under overhangs at the foot of the rock formations known as Alindzha-Dag, 
Ilyanlu-Dag and Darry-Dag, and in caves near the village of Klit, east of 
Ordubad, at the beginning of the Araks gorge. 

The present-day landscape of the Abrakunis area consists of plateau-like 
spurs, covered with spiny astragalus bushes, mountain saltwort and weakly 
developed ephemeretum. The relict shrub vegetation, which had 
persisted in the valleys, consists of almonds, buckthorn and ephedra. This 
vegetation indicates that a xerophytic landscape existed in the region even 
before cattle husbandry. The plateau-like spurs are cut by dry canyons with 
rock-covered areas and caves, which are the resting sites of owls and 
eagle owls. There has been a survey of one such cave in the "Hyena ravine" 
Kyaftar-Dara (Figure 105). 

On the left wall of the Araks gorge we surveyed three karst caves located 
between the Ordubad and Megri stations near the village of Klit. 

These caves were visited in 1879 by Polyakov (1882) on his reconnaissance 

229 of Paleolithic sites. In a trench in the anterior chamber of the middle cave, 
at a depth of 1.5 m, Polyakov found two pelvic bones of wild goat (C a p r a 
aegagrus), leg bones of European brown bear (U r s u s arctos), several 
bones of hare (Lepus europaeus), "femur of a small carnivore similar 
to a dog" and bones of a pigeon. The excavations revealed the following 
sequence from topto bottom: a layer of flaky clay and excreta of bats, 35 cm 
thick, clayey beds with excreta and bones of bats. 

All three caves are located in a wall of dolomitized limestone on the 
eastern slope of one of the small valleys in the Araks ravine. There are 
other caves in this limestone which, however, are not easily accessible. 
The biggest cave is the middle stalactitic cave "Dlinnokryly", described by 
Satunin (1915a). The entrance slopes downward into a cavern, from the 
middle of which the floor slopes upward to a long, narrow corridor with 
calcitic encrustations. From the posterior part of the cavern another 
twenty-nneter-long corridor leads to the left into a lower chamber, 15 m 
wide, and to yet another corridor, this one proceeding a length of 50 m to 
a dead end. The combined length of the cavern and the long corridor is 250 m. 

230 The height of the first chamber is 12-18 m; some sections are higher. 
In the talus left by Polyakov's excavations bones of Miniopterus 
schreibersii were observed. In the left corner of the cave, near the 
resting place of some carnivore (panther?), we found 14 gnawed bones 
(horse, ass, sheep, tur) showing a recent mode of preservation, and 
several bones of Miniopterus schreibersii covered with calcitic 
crust. Bones of a young panther (Panthera pardus) and stone marten 
(Martes foina), also showing a recent type of preservation, were found 
in the stone -covered middle section of the cave. 

Of the present cave inhabitants, we caught only one greater horseshoe 
bat (Rhinolophus ferrum e qu i nu m) and counted only ten flying 
Miniopterus. Satunin in 1893 counted Miniopterus in the thousands. 
Footprints cf panthers, martens and white -toothed shrews occur in the 
ancient dust. 



226 



[229) 




FIGURE 105. Entrance to cave inhabited by eagle owl in Kyaftar-Dara ravine 

Photograph by author, 1947 

Some rodent bones from the pellets of eagle owls were collected near 
the mouths of the two marginal caves. 

Comparison of Sosnikhina's collections of small species with ours 
indicates a considerable increase — from west to east — in the xerophilous 
fauna and a concomitant species impoverishment in the small-mammal 
complexes (Table 59). 

All of this material from the Arals valley — up to 1,500 bones from owl 
pellets — was collected from the surface and generally shows a recent type 
of preservation. It gives no indication of the appearance or the disappearance 
of any species in the semideserts in our time. A similar composition of the 
fauna was established through sampling with traps. However, in the 
interests of accuracy, it is necessary to mention that a mouse-like hamster 
was discovered as recently as 1938 on Mount Darry-Dag near Ilyanlu-Dag, 
and that bones have been found of Asia Minor hamster which does not inhabit 
the upland semidesert zone at present. 



231 Talysh and El'brus upland 

No bones of the postglacial epoch are known from the northern foothills 
of the Talysh Mountains and the Lenkoran Plain; neither are food remains 



227 



known from the Talysh highland steppe. However, ancient burial mounds, 
in southern Mugan, as yet untouched by archaeologists, are evidence of 
ancient human habitation of the region. 



(230) 



TABLE 59. Incidence of species (in'/o)in mammal remains from pellets of owls and eagle owls in caves 
of the middle Araks Vallev 



Species 


Novyi Shakhvarut, 
Ml, Karakhan 


Abrakunis, 

Kyaftar-Dara 

cave 


Ordubad, 

Dlinnokryly 

cave 




25.2 
1.2 
0.6 
0.6 
3.7 
4.9 

14,2 
4.3 
0.6 

14.2 

0.6 
4.3 

1.2 
17.8 

0.6 


3.1 
3.1 

6.4 
6.4 

3.1 
6.4 

1 40.0 / 

6.4 
6.4 
3.1 

6.4 
9.2 






_ 


Crocidura russula 


_ 


Miniopterus schreibersii 

Mustela nivalis 


- 


Lepus europaeus 


6.7 


Allactaga williamsi 




Allactaga elater 


_ 


Mus musculus 


6.7 


Apodemus sylvaticus 




Meriones blackleri 




M.persicus 


80.0 


Mesocricetus auratus 


6.7 


Cricetulus migratorius 




Calomyscus bailwardi 




Arvicola terrestris 


_ 


Microtus arvalis 




M. socialis 


_ 


Ellobius lutescens 









In 1945 I collected rodent bones in small caves inhabited by owls located 
in three sections of upland steppe, covered with grass and spiny astragali, 
in the upper reaches of the Vilyazh-Chai, near the Dymanskii post and in 
the upper reaches of Vassaru-Chai, near Kelakhan and Kel'vyaz. 



TABLE 60. Species and number of mammal bones from owl pellets in upland steppe of Talysh 



Species 


Number of bones 


Number of individuals 


Hemiechinus auritus 


2 

1 
53 

2 

35 

111 

76 

1 
13 


1 


Lepus europaeus 


1 


Allactaga williamsi 


9 


Meriones persicus 


1 


Cricetulus migratorius 


14 


Mesocricetus auratus 


17 


Microtus socialis 


15 


M. arvalis 


1 


Ellobius lutescens 


7 






Total 


294 


66 







228 



The age of these bones is not more than 200-300 years. In most cases, 
they were not protected by soil and consequently were subject to weathering. 
The combined data for all the three sections are given in Table 60. 

All the small mammals listed still inhabit the region, forming the Recent 
assemblage characteristic of the upland steppes of the northwestern Iranian 
Plateau and the Armenian Highland. The list does not include the European 
hedgehog, long-tailed shrew, steppe mouse and snow vole. These species 
are very rare in the region. (Relict snow voles occur higher in the 
mountains. ) The similarity of our list to Sosnikhina's (1947, 1948) list, 
given for the tragacanth-covered steppe areas of the Armenian Highland, 
proves a considerable degree of affinity between the faunas and between the 
landscapes of the two regions. 

On the adjacent Iranian Plateau, Coon (1951) collected Holocene 
mammials from the eastern part of the Elburz Range, 50 km from Asterabad, 
in Mesolithic and Neolithic beds of the Belt cave. Among the 1,170 fragments 
of bones, 18 species have been identified. The most abundant is goitered 
gazelle (Gazella subgutturosa, 67%), followed by Bos 
primigenius (18.8%), goat (C a p r a sp. , 5.7%), ram (O v i s sp. , 
4.4%), Caspian seal (Phoca caspica, 3.1%) and dog (Canis sp. , 
1.7%). Also mentioned are individual occurrences of Vulpes sp. , Felis 
sp. , Viverridae, Citellus sp. , Castor fiber, Hystrix sp. , 
Mus sp., Equus asinus, Sus sp., Cervus elaphus and Cervidae. 
232 Identification of the bones of the Viverridae, European beaver and Bos 

primigenius needs to be confirmed. The remaining species are to some 
extent similar to the present assemblages of mammals on the hot plateaus 
of northwestern Iran and humid, wooded Asterabad. 

These studies of the various types of Holocene deposits of the 
Armenian Highland, Araks valley, and Talysh and El'brus highlands show 
that a number of large animals (bison. Bos primigenius, kulan, 
goitered gazelle) inhabited the region in the not-distant past. They also 
reveal shifts in the areas of distribution of some smaller species (suslik 
and Asia Minor hamster). 



The results of the studies of Holocene localities on the Caucasian 
Isthmus can be summarized as in Tables 62-63. 

In our studies of the bone collections from the post-Paleolithic cultural 
beds and from diluvial sediments, we were able to trace the appearance 
of some species of large mammals in the Caucasus in postglacial time. 
The species migrated to the Caucasus from adjacent areas. Extinction of 
many of the Pleistocene species was also revealed by the studies. The 
rapid development of animal husbandry and its characteristic regional 
features, as well as the steady increase in the proportion of domestic over 
wild species in food remains from the Neolithic to the present, could be 
traced through the bone material. Studies of bone material from the day 
resting places of eagle owls produced a more precise delineation between 
past and present distribution of many insectivores and rodents. The rarity 
of multilayered Holocene deposits of small bones and the abundance of 
younger bones in small caves and under rocky ledges might indicate a 
discontinuity in the availability of shelters and in the accumulation of food 
remains of owls. It is possible, however, that rapid mechanical weathering 



229 



of *he exposed rodent bones and the frequent collapse of small caves might 
have created the appearance of such a discontinuity. 

The studied Holocene material consists of 77 species of mammals which 
inhabited the Caucasus at the end of the Upper Paleolithic, some of which 
became extinct toward the 20th century. 

Comparison of the lists of the Holocene and Pleistocene species indicates 
pronounced changes in the compositions and distribution of the mammalian 
fauna of the Caucasus, occurring mainly at the Pleistocene-Holocene 
boundary (Table 61 ). 

Thus, the Holocene in the Caucasus is characterized by the complete 
disappearance of proboscideans* and apes while perissodactyls and 
artiodactyls decreased catastrophically. The increase in the number of 
rodents in the Holocene beds is accounted for by the differences between 
excavated and open sites. 



TABLE 61. Number of mammalian species from the Pleistocene and the Holocene in the Caucasus 



Species 


Pleistocene 


Holocene 




6 
2 
1 

21 
1 

26 
2 
6 

18 


9 




3 


Primates 


_ 




14 


Lagomorpha 


2 
33 


Proboscidea 


_ 


Perissodactyla . . 
Artiodactyla . . , 




2 

14 




Total 


83 


77 









Reliable conclusions on the appearance or disappearance of species in 
the Holocene can be drawn only with respect to the mass species. Overall, 
the species complexes and locations of the Holocene deposits exhibit a nearly 
modern faunal composition and distribution pattern. 
233 A review of paleontological material on post-Pleistocene fauna shows 

the following changes: 

Insectivora — At least five species crossed the Pleistocene -postglacial 
boundary: mole, white-toothed shrew, water shrew, long-eared and 
European hedgehog. In addition, three species of white-toothed shrew 
(Crocidura suaveolens Pall., С r о с i du r a leucodon Hermes, 
Suncus etruscus) occur in the Holocene; these species probably inhabited 
the Caucasus in the Pleistocene also. 

Chiroptera — Greater horseshoe bat (R h i no lo ph u s) is known from 
the Pleistocene beds. The Holocene beds in the caves of the Lesser 
Caucasus uplands commonly contain bones of M yot i s and Miniopterus. 

Primates — There are no data on the occurrence of Holocene macacas. 

" In view of the finds of geologically young mammoths in Pyatigor'eandin the Gori depression, the question 
of their possible survival through the Pleistocene in the Caucasus deserves further investigation and requires 
additional corroborative material. 



230 



Carnivora — Of the 21 species known in the Pleistocene, only 15 have 
been recorded in the Holocene. In the Holocene of the Caucasus there are 
no steppe or cave bear, cave hyena, Panthera leo or cheetah and probably 
no glutton. Jackal and striped hyena occur for the first time in the Holocene. 
It seems possible that at this period a number of other southern carnivores 
(jungle cat, lion and tiger) naigrated to the Caucasus although their remains 
have not yet been found. The absence of lion, tiger and cheetah bones from 
postglacial deposits can be accounted for by their scarcity in the historical 
epoch. 

Lagomorpha — European hare is the only species known from the 
Pleistocene. The Holocene deposits contain remains of European hare and 
large pika which probably existed in the Lesser Caucasus in the Pleistocene. 

Rodentia — A number of important changes in the composition and 
distribution of the rodent fauna occurred during the Holocene. The porcupine 
and the Apsheron vole became extinct; the Asia Minor hamster retreated 
fronn the Apsheron and from Imeretia; the areas of distribution of the suslik 
of Asia Minor on the Armenian Highland and of the Caucasian snow vole in 
Ciscaucasia decreased. The finds of renaains of black rats in Neolithic beds 
on the Black Sea coast and in Kura Plain beds of the middle of the first 
244 millennium B.C. are a basis for assuming that the species was autochthonous 
and that it survived through the Pleistocene on the warm, humid lowlands 
near the Black and southern Caspian seas. A similar conclusion, if applied 
to the Norway rat, would be open to question. The occurrence of European 
beaver in Neolithic times in Colchis has been confirmed. 

Most of the other findings fill in details of the distribution of extant 
species. 

Perissodactyla — The materials studied give no indication that the woolly 
rhinoceros and the Equus hemionus survived in the Caucasus through 
the Pleistocene. The occurrence of the modern kulan in the Holocene of 
eastern Trans- and Ciscaucasia has been recorded. There were undoubtedly 
wild horses in the Holocene, but it has not yet been possible to distinguish 
their bones from the bones of dom^estic horses of various origins. Wild and 
domestic horses were the main animals used in sacrifices and for food by 
the Caucasian tribes during most of the historical epoch. 

Artiodactyla — Boar were widely distributed in the Caucasus. In the 
Holocene the species was one of the primary game animals. Bos 
primigenius were also abundant in the Lesser Caucasus. So far there 
is no evidence of the existence of wild camels in the Caucasus in the 
Holocene, 

The wide distribution of red deer has been traced over nearly the entire 
Caucasian Isthmus from the Pleistocene to the 19th century. Roe deer 
and elk inhabited the lower Don valley, the river valley forests and reed 
jungle of the Kuban and the wooded lowlands of Ciscaucasia and Colchis to 
the end of the 18th century. No Holocene remains of Cervus 
pliotarandoides and fallow deer are known . 

The occurrence of bison has been established for the Bronze Age in 
the Lesser Caucasus, and for the Middle Ages of the present era on the 
middle and lower Don. Saiga disappeared from eastern Transcaucasia in 
the Lower Holocene. Remains of goitered gazelle begin to occur in beds' 
dated second m.illennium B.C. in eastern Cis- and Transcaucasia. They 
are particularly abundant in the Apsheron Peninsula beds of the Middle Ages. 



231 



234 '^'^BLE 62. Species and stratigraphic location of mammal bones from the Quaternary of the Caucasus 





Pleistocene 


Holocene 


Orders and species 


О 


2 


a. 
a, 
D 


(J 
о 

«s 


о 

XL 

о 
u 
Z 


DO 

< 

с 
о 


СЮ 

< 
о 


DO 
< 

-о 


0^ 

с 

0) 

a. 


Insectivora 




















Erinaceus europaeus L. (s. lato) 


- 


+ 


+ 


- 


- 


+ 


+ 


+ 


+ 


Hemiechinus a uri tus Gmel. (s. lato) 


- 


+ 


- 


- 


- 


+ 


- 


+ 


+ 


Sorex minutus L 


- 


- 


- 


- 


- 


- 


- 


- 


+ 


S.araneus L 


- 


- 


- 


- 


- 


- 


- 


- 


+ 


Neomys fodiens Schr. (s. lato). . . . 


- 


+ 


- 


- 


- 


- 


- 


+ 


+ 


Crocidura leucodon Herm 


- 


- 


- 


- 


- 


+ 


- 


+ 


+ 


C, russula Pall. (s. lato) 


- 


+ 
+ 


+ 


- 


- 


+ 


+ 


+ 
+ 

+ 






+ 


Talpa caucasica Sat 




T- sp 




Chiroptera 




















Rhinolophus f e r ru m - equ i n u m Schr. 


- 


- 


+ 


- 


- 


- 


- 


- 


+ 


Myotis oxygnathus Mont 


- 


- 


- 


- 


— 


- 


- 


+ 


+ 


Miniopterusschreibersii KUhl. . . 


- 


- 


- 


- 


+ 


+ 


+ 


+ 


+ 


Chiroptera gen 












_ 














Primates 






















+ 


+ 














_ 


Camivora 




Canis aureus L 


+ 


+ 
+ 


+ 


: 


+ 


+ 


- 


+ 


_ 


C. lupus L. (s. lato) 


_ 


C. lupus apscheron i cus N. Ver. . . 


- 


"C. tamanensis N. Ver 


+ 


- 


- 


- 


- 


- 


- 


- 


- 


С a n i s sp 


- 


+ 
+ 
+ 
+ 
+ 


+ 


- 


+ 


+ 


+ 


+ 


_ 


С u П sp 




Vulpes corsac Pall 




V. vulpes L. (s. lato) 


+ 


V. vulpes aff. alpherakyi Sat. . . . 




и rsus a re tos L 


- 


- 


+ 


- 


+ 


+ 


+ 


- 


- 


U. arctos binagadensis N. Ver. . . 


- 


+ 


- 


- 


- 


- 


- 


— 


— 


U. spelaeus Rosm.(s. lato) 


+ 


+ 


+ 


- 


- 


- 


- 


- 


- 


U. rossicus Boris 


— 


+ 


— 


~ 


+ 




+ 




_ 


Hyaena hyaena L 


- 


Crocuta spelaea Goldf 


+ 


+ 


+ 


- 


- 


- 


- 


- 


- 


Martes foina Erxl 


— 


— 


: 


— 


+ 


+ 
+ 


+ 


+ 


+ 


M. foina latifrons Sat 




M. cf. foina Erxl 





+ 


+ 
+ 


— 





_ 





— 


_ 


M. cf. martes L 


- 


Vormela peregusna Guld 


- 


+ 


- 


- 


- 


+ 


+ 


- 


— 


„or Lutra lutra L 


+ 


- 


+ 


- 


- 


- 


+ 
+ 


- 


_ 


235 

Mustela nivalis L. (s. lato) 


- 



* Species of the transitional Upper Pliocene-Lower Pleistocene period on the Taman Peninsula are marked 
with an asterisk. 



232 



TABLE 62 (continued) 



Orders and species 



Pleistocene 



Meles meles L. (s. lato) 

M. melesurartuorumSat 

M. meles aff. minor Sat 

Meles sp 

Gulo aff. gulo L 

Felis silvestris Schr. (s. lato) . . . 

F. lybica Forst 

F. aff. lybica Forst 

F. lynx L. (s. lato) 

Panthera pardus L 

P. spelaea Goldf 

'■' P a n t li e r a sp 

Acinonyx aff. jnbatus Schr 

Lagomorplia 

C) с h о t о n a sp 

Lepus europaeus Pall 

L. europaeus gureevi I. Groni . . . 
L, cf. europaeus 

Rodentia 

M a r in о t a sp . 

Cite 11 us pygmaeus Pall 

C. aff. citellus xanthoprimnus 

Benn 

Castor fiber L 

* C. tamanensis N. Ver 

* Trogontherium cuvieri Fisch. 

Hystrix of. leucura Sykes 

H. V i П о gr a d о V i Arg 

Dyromyp nitedula Pall 

G 1 is gl i? L 

Sicista sub til is Pall 

S. cf. caucasica Vin 

Allactaga jaculus Pall 

A. jaculus bogatsciievi Arg. . . . 

A. Williams i 'I'lios 

A. Williams i dzhafarovi I. Grom. 

A . cf. w i 11 ia m s i 

A , e lat e r Licht 

Spalax microphthalmus Giild . 

S. giganteus Nehr 

Micromys minutus Pall 

Mus musculus L.(s. lato) 

Apodemus agrarius Pall 

A.sylvaticus L. (s. lato) 

Rattus rattus L 

R. norvegicus Berk 



233 



236 TABLE 62 (continued) 





Pleistocene 


Holocene 


Orders and species 








'•J 




ОС 




0) 
00 


Е 






i 


C- 
C 


5 


"o 

0) 

Z 


< 

N 

О 


< 

с 
о 


< 

■о 
-а 

2 


с 
1) 

Си 


Cricetus cricetus L 


- 


- 


- 


- 


- 


+ 
+ 


+ 


+ 
+ 


+ 


Mesocricetus auratus Water 


+ 


M. aff. auratus Water 


_ 


_ 


_ 




: 


_ 


+ 


+ 


_ 


M. auratus raddei Nehr 


+ 


M. auratus planicola Arg 


- 




- 


- 


- 


- 


- 


- 


- 


Cricetulus rnigratorius Pall 


- 


- 


- 


- 


- 


+ 


+ 


+ 


+ 


C. rnigratorius argiropuloi I. Grom . 


- 




- 


- 


- 


- 


- 


- 


- 


Calomyscus bailwardi Thos 


- 


- 


- 


- 


- 


- 


- 


— 


+ 


Meriones erythrourus Gray 


- 


- 


- 


- 


- 


- 


- 


+ 


+ 


M. erythrourus intermedius I. Grom. 


- 


+ 


- 


- 


- 


- 


- 


- 


- 


M. blacl<leri Thos 


- 


- 


- 


- 


- 


- 


+ 


+ 


+ 


M persicus Blanf. 




_ 


: 


: 


: 


_ 


+ 


+ 


+ 


Lagurus lagurus Pall 


+ 


Arvicola terrestris L. (s. lato) 


- 


■^ 


- 


+ 


- 


+ 


+ 


+ 


+ 


Microtus arvalis Pall 


- 


+ 


- 


- 


- 


- 


- 


- 


- 


M. socialis Pall 




+ 


4- 


: 





+ 


+ 


+ 
+ 


^ 


M. cf. roberti Thos 


- 


M. nivalis Mart 


+ 

+ 


+ 


- 


+ 


- 


+ 


+ 


+ 
+ 


+ 


M aff gud Sat 




M aff majori Thos 


_ 







Prometheomys aff. schaposchnikovi 




Sat 


+ 


: 


+ 


+ 


: 


: 


_ 


_ 


— 


Ellobius talpinus Pall 


+ 


E. lutescens Thos 


- 


- 


- 


- 


- 


- 


+ 


+ 


+ 


E. aff. lutescens Thos 


- 


+ 


- 


- 


- 


- 


- 


- 


- 


Muridae gen 


+ 


- 


- 


— 


- 


— 


— 


— 


— 


Proboscidea 




















"Elephas meridionalis Nesti 


X 


- 


- 


- 


- 


- 


- 


- 


- 


"E.antiquus Falc 


+ 


- 


- 


- 


- 


- 


- 


- 


- 


E. trogontherii Pohl 


+ 


+ 


- 


- 


- 


- 


- 


- 


- 


E. primigenius Blum 






+ 


+ ? 










— 


Perissodactyla 




Equus hemionus Pall 


- 


- 


- 


- 


- 


+ 


+ 


+ 


- 


E.aff. hidruntinus Reg 


- 


+ 
+ 


+ 

+ 


+ 


Not с 


+ 


+ 


+ 

от do 





E, caballus L. (s. lato) 


+ 


E. caballus gmelini Ant 


listinguished fr 


mestic 














horses 






" E. aff. sussenbornensis Wiist 


+ 


- 


- 


- 


- 


- 


- 


- 


- 


* E. stenonis Cocchi 


_ 


_ 


+ 


~ 


■~ 


_ 


: 


_ 


— 


Rhinoceros tichorhinus Fisch 


- 


Rh. cf. mercki Jaeger 


+ 
+ 


+ 
+ 
+ 


- 


- 


- 


- 


- 


- 


— 


Rh. binagadensis Dzhaf 


_ 


Rh. sp 


- 



234 



237 TABLE 62(continued) 





Pleistocene 






Holoc 


ene 






Orders and species 






0) 

с. 


о 
"о 


о 

"о 
z 


to 

CO 

< 

N 

с 

2 

03 


DO 

< 

О 


00 

< 

a 

-a 
-a 

2 


В 

с 
u 

D- 


* Rh . cf. etruscus Falc 

Elasm otherium sibiricum Fisch. . . . 
*E. caucasicum Boris. 


+ 
+ 

+ 

+ 
+ 

+ 
+ 

+ 
+ 

+ 
+ 

+ 
+ 

+ 
+ 
+ 

+ 
+ 

+ 
+ 
+ 


+ 

+ 
+ 

+ 

+ 

+ 
+ 

+ 
+ 

+ 

+ 

+ 
+ 

+ 

+ 


+ 

+ 

+ 

+ 

+ 

+ 
+ 

+ 

+ 

+ 
+ 

+ 

+ 
+ 


+ 
+ 

+ 
+ 
+ 

+ 


+ 

+ 
+ 

+ 
+ 

+ 

+ 
+ 


+ 

+ 

+ 

+ 
+ 
+ 

+ 

+ 
+ 


+ 

+ 

+ 
+ 
+ 

+ 
+ 


+ 

+ 

+ 

+ 
+ 

+ 
+ 
+ 

+ 
+ 


- 


Artiodactyla 
Sus scrofa L. (s. lato) 


+ 


S. apscheroni cus Bur. et Dzhaf 


- 




_ 


Camelus dromedarius dahliChav. 
(ferus ? ) 




С knoblochi Nehr 





*Paracamelus gigas Schloss 

'*P. cf. kujalnikensis Chom 

Cervus elaphus L. (s. lato) 

*Eucladocerus sp 

*Cervidae gen. . , 


+ 


Dama cf. mesopotamica Brooke . . . 

Megaceros euryceros Aldr 

'Megaceros sp 


— 


Alces alces caucasicus N. Ver. . . . 
A. aff. alces L 


+ 


*Tamanalces sp 





Capreolus capreolus L. (s. lato) . . . 
Saiga tatarica L 


+ 


Gazella subgutturosa Guld 

*Gazella sp 


- 


*Tragelaphussp. . . 


_ 


*Strepsicerotini gen. et sp. 

Rupicapra ruplcapra caucasica 
Lyd 


+ 


Capra cylindricornis Blyth 

C. caucasica Giild 


+ 

+ 


C. aff. caucasica Giild 





C. aegagrus Gmel 


+ 


Ovis gmelini Blyth 


- 


0. cf, ammon L 


_ 


Bos minutus Malsb 





B. primigenius Boj 





B, trochoceros Leitn 





B. mastan-zadei Burtsch 


_ 


Bison bonasus caucasicus Sat. ... 
B. priscus Boj. (s. lato) 


+ 


'B. cf. schoetensack i Freud 

B. sp 

* Bison and Bos 


- 







235 



238 "^ABLE 63. Stratigraphic summary of main localities of Quaternary mammalian index species in the 





,„ «f 


Marine 
beds 


Black Sea region 


Climate and landscape 
from the composition 
of mammalian fauna 


Quaternary 


Western Ciscaocasia 
and Russian Plain 


Western Transcaucasia 










Sarkel fortress, 




Recent climate and 










9- 13th centuries 




increasing influence 










European brown bear 




of man on landscape 




t~i 




European beaver 








a. 


с 
a. 

u 

< 
с 
о 
t-. 

<u 

DO 




kulan 

elk 

red deer 

saiga 

bison 

CI- 28^ 

Semibratnoe site at 
Kuban estuary 
wolf 
otter 
boar 
deer 


Huts of Colchidians at 


Development of broad- 
leaf forest in foothills 


с 

(U 

о 

_o 

о 

X 


•o 

is 

о 


< 

О 
Ш 

"о 
u 
Z 

о 
2 


с 
u 

E 
u 

с 

(U 

о 

0) 

ей 


Mud-hut settlement at 
Gelendzhik 
lynx 
hare 
red deer 
dolphin 
CI- 41 7 


Anaklia and Ochamchiri 
jackal 
boar 
roe deer 
red deer 

Akhshtyrskaya cave, 
upper beds 
European brown bear 
badger 
black rat 
roe deer 
CI- 460 


Strong warming and 
development of steppes 
in foothills; waning of 
valley glaciers 



Calcination indexes of bones (CI) by collagen analysis (Pidoplicnko, 1952); analyses aone in I.G. 



236 



239 Caucasus and adjoining lands 



Marine 
beds 



Caspian region 



Eastern Ciscaucasia and 
Russian Plain 



Dzuars of North Ossetia 
15- 19th centuries 
red deer 
roe deer 
Caucasian elk 
east Caucasian goat 
chamois 
Caucasian bison 



Settlements of Sunzha 

valley, bti-Su, Alkhan- 
Kala 

kulan 

red deer 

Settlements at Kayakent and 
Dzhemikent 
fox 
bear 
kulan 

goitered gazelle 
CI- 275 



Eastern Transcaucasia 



Settlements near Baku, 
9- 13th centuries 
fox 
kulan 

goitered gazelle 
seal 



Burials in jugs at Mingechaur 
and settlement at Sumgait 
striped hyena 
black rat 
boar 
red deer 
goitered gazelle 



Sosruko grotto on the 
Baksan 
hare 

little suslik 
boar 
red deer 
Caucasian goat 
CI- 241 

Chokh site in Dagestan 
hamster 
mouflon 
CI-317 



Lesser Caucasus upland and 
northern Iranian Plateau 



Karmir-Blur fortress 
stone marten 
Armenian mouflon 
wild goat 
goitered gazelle 



Pile dwellings and burials on 
Sevan shores 

stone marten 

badger 

boar 

red deer 

dwarf tur 

CI- 207 
Belt cave at Asterabad 

porcupine 

red deer 

goitered gazelle 

Bos primigenius 

seal 



Pidoplichko's laboratory in Kiev. 



237 



240 TABLE 63 (continued) 



Subdivision of 
Quaternary 


Marine 
beds 


Black Sea 


region 


Climate and landscape 
from the composition 
of mammalian fauna 


Western Ciscaucasia 
and Russian Plain 


Western Transcaucasia 










Bone-bearing alluvium 














of late terraces of 














Don 














fox 


Gvardzhilas cave 


General strong cooling 




5 

a. 


о 

(L) 




small pika 
little suslik 


mole 

European brown bear 


and probable displace- 
ment downslope of 








3 
U4 


great jerboa 


glutton 


landscape zones 






(U 

a. 


О 


water vole 


Asia Minor hamster 


following glaciation of 






a. 


Z 

nj 
DO 


horse 

elk 

red deer 

reindeer 

saiga 


Prometheomys 
Caucasian goat 
bison 
CI-512 

Sakazhia, Devis-Khvreli, 
Mgvimevi caves 
cave bear 


high ridges 


<u 






nj 




horse 


Major uplift of the 


<u 






^ 




European beaver 


mountain ranges 


о 










elk 




Q) 










chamois 




a. 






О 

-a: 

CD 

с 

N 
D 


Il'skaya site at Krasnodar 
cave hyena 
mammoth 
boar 
red deer 
saiga 
Bison priscus 


Caucasian goat 

argali 

CI-580 


Altitudinal zonation of 




2 




с 
3 


CI-518 
Bone-bearing alluvium 


Akhshtyrskaya cave, 


modern type 






■o 


1> 


of early terraces of 


lower beds 








2 

1 


< 


Don 
Siberian polecat 
little suslik 
jerboa 
Allactaga sibiri- 

ca saltator 

Eversmann 
S pa 1 a X 
red deer 


wolf 

cave bear 
common hamster 
red deer 
giant deer 
Bison priscus 
CI-706 


" 



238 



241 





Caspian region 




Marine 
beds 






Lesser Caucasus upland and 
northern Iranian Plateau 


Eastern Ciscaucasia and 
Russian Plain 


Eastern Transcaucasia 




Diluvium in Yankul' gulley 




Diluvial loams at Maragheh 




on upper Kalaus 




wolf 




horse 




cave hyena 




saiga 




horse 

rhinoceros 

argali 






Bituminous loams Kir-Maku 


Tamtama and Bisotun caves in 






and Artem I, 


northern Iran 




Upper travertine complex of 


wolf 


jackal 




Mt. Mashuk in Pyatigor'e 


horse 


cave hyena 




mammoth 


red deer 


panther 




horse 


saiga 


gerbil 


0) 
00 


roe deer 


European tur 


red deer 


2 




CI 158-208 


goitered gazelle 


л 








с 






Zurtaketi site on upper Khram 








horse 
ass 




Bone-bearing sands of Terek, 




mouflon 




terraces at Mozdok 


Bituminous sands at Binagady 


Bison priscus 




Bison priscus 


on Apsheron Peninsula 
long-eared hedgehog 
corsac fox 
Ursus arctos binaga- 

dens is 
tiger polecat 
cheetah 
red- tailed gerbil 




ьо 
a 

N 
nj 

л: 


Bone-bearing alluvium of 
ancient extended Volga 
valley 

cave hyena 

Panthera leo 

mammoth 

woolly rhinoceros 


porcupine 

Rhinoceros binaga- 
densis 
red deer 
saiga 
tur 
CI 450-520 




horse 








Camelus knoblochi 








giant deer 








long-horned bison 








CI-484 







239 



242 



TABLE 63 (continued) 



Subdivision of 
Quaternary 


Marine 
beds 


Black S 


ea region 


Climate and landscape 
from the composition 
of mammalian fauna 


Western Ciscaucasia 
and Russian Plain 


Western Transcaucasia 










Bone-bearing sands at 




In Transcaucasia strong 










Girei and Krasnodar, 


Kudaro cave on upper 


development of steppes 










middle beds 


Rion, lower beds 


on the plateaus. 










Elephas trogon- 


macaca 


resulting in migration 










t he ri i 


Siberian red dog wolf 


of southwest Asian 










long-horned bison 


cave bear 


species to Caucasus 










bear 


panther 

Asia Minor hamster 

marmot 




с 
u 

о 

'S 

Cl 


о 


о 

a 

a. 

% 
о 

J 


и 


Bone-bearing sands at 
G irei and Ivanovskaya, 
on Kuban, lower beds 

southern elephant 

Elephas antiquus 

deer 

bison 

Conglomerates and sands 
at Kuchugury, Sinyaya 
gulley and Tsimbal 
on Taman Peninsula 
Trogontherium 

cu vier i 
Castor tamanen- 

sis 
southern elephant 
Elephas antiquus 
Elasmotherium 
caucasicum 


P гоше th eomys 

vole 

porcupine 

rhinoceros 

boar 

roe deer 

argali 

CI- 914 


On Ciscaucasian plains 
climate temperate and 
dry; broadleaf forests 
in foothills 


lU 

с 

о 

о 

a. 


V 

о, 

Cl. 

D 


"o 


J3 

i 

с 

я 


horse 

Cervus pliota- 

randoides 
camel 
antelop 
bison 
CI 913-1203 

Sands and gravels of ancier 
terraces of Kuban at 
Voskresenskaya and 
Psekupsat Babinskaya 
andSaratovskaya 
southern elephant 
Equus stenonis 
Rhinoceros etrus- 

cus 
Cervus pliotaran- 

d ides 
CI- 1033 


Fauna unknown 
t 


On Ciscaucasian plains; 
warm climate; 
savannah and 
gallery forests 



240 



243 





Caspian 


region 




Marine 
beds 






Lesser Caucasus upland and 
northern Iranian Plateau 


Eastern Ciscaucasia 








Eastern Transcaucasia 






and Russian Plain 












Bone-bearing sands of Adzhi- 








Eilas near Yerevan 








Elephas trogontherii 








fallow deer 








tur 








Paleolithic sites on middle Zanga 








and at the foot of Alagez; 








fauna unknown 








Bone-bearing sands at Leninakan, 








upper beds 








Elephas trogontherii 








horse 


u 






Rhinoceros mercki 


so 








Э 






giant deer 


a 






Camelus knoblochi 


^ 






Bos pr i migenius 




Bone- bearing sands of upper 


Marine coquina at Khurdalan 






terraces of Kuma and 


on Apsheron Peninsula 






Podkumok near Georgievsk 


horse 






southern elephant 


rhinoceros 






Equus stenonis 








deer 








Bos 








CI 1073 


Sub-Khazar beds at Binagady 
Elasmotherium 
s ib iri cum 

Marine coquina at Kishla on 
Apsheron Peninsula 
hyena 


Bone-bearing sands at Leninakan, 
lower beds 
Mastodon arvernensis 




Bone-bearing sands at 


Equus stenonis 






Podlesnyi and Divnyi 






о 


near Manych 


Diluvium at Tsinandali 




1 


southern elephant 


Rhinoceros etruscus 


Nurnus on Zanga 


< 


Elasmotherium 




small Mustilidae 


П) 


caucaslcum 






CL 






hare 


D 






Rhinoceros etruscus 




Tash-Kala near Grozny 


Yenikent on Alazan 


Oryx 




Elephas planifrons 


Trogontherium 






Equus stenonis 


cu vieri 





241 



TABLE 63 (continued) 







Black Sea region 


Climate and 


Subdivision of 


Marine 




landscape from 






Quaternary 


beds 


Western Ciscaucasia and 
Russian Plain 


Western Transcaucasia 


the composition of 
mammalian fauna 










Bone- bearing sands at 














Khapry and Rostov on 




Warm climate; 










Azov Sea coast 




savannahs and 










Elephas planifrons 




gallery forests in 








^ 
^ 


Mastodon arvernensis 


Fauna unknown 


the southern part 


4J 




О 


x; 


H ippa rion 




of the Russian Plain 


С 
О 


(L) 


-£ 


-o 


Equus stenonis 








D- 


Ш 


5 


camel 






a. 






2 


CI 1066-1142 







245 Paleontological studies of Holocene mammals lead to a number of general 

conclusions on faunal and landscape changes in the Caucasus which have 
occurred since the end of the Upper Pleistoce. The absence from the Holocene 
beds of those mammals which lived during the stage of maximum cooling indi- 
cates that a discontinuity in ecological conditions promoted their extinction. 

The paleontological materials indicate that the ecological assemblage of 
the piedmont plains was the one most affected in postglacial time; mountain 
fauna was probably affected to a much lesser degree. 

Inasmuch as species such as cave bear, cave hyena, Panthera leo, 
Equus hemionus, giant deer and others lived in the cold climate of 
the Upper Pleistocene, it may be assumed that the main reasons for their 
extinction were abrupt warming and increasing drought. The development 
of xerothermal landscapes must have affected the fauna of the lowlands 
and foothills to the east much more pronouncedly than the fauna in the 
western part of the Caucasian Isthmus. A number of relevant examples of 
the retreat of mesophilous species have been given under the discussion 
of the Binagady locality. 

The Holocene deposits, like the Pleistocene, do not contain any northern 
cold-climate forms (reindeer, blue hare, arctic fox). No Holocene remains 
of these species were found on the lower Don either. It becomes clear that 
in the Holocene the ranges of distribution of these species moved north, 
rather than closer to the Caucasus. Development of isolated pockets 
inhabited by mountainous mesophilous rodents also indicates the advent of 
a warm climatic phase, which resulted in shifts in the landscape zones. 
Examples of such isolated pockets are: pine vole on the meadows of the 
Stavropol Plateau, Transcaucasian mole vole on Mount Razvalka in Pyatigor 'e, 
Asia Minor snow vole on Mount Kelakhan in the dry Zuvanda valley of the 
Talysh Mountains, and steppe vole on the Apsheron Peninsula. These 
and other examples of disrupted ranges of distribution are reviewed in 
more detail in a separate section under the discussion of the history of the 
development and the present state of areas of distribution of some species. 

The new developmental stage in Upper Quaternary faunal complexes is 
marked by the appearance in the Caucasus of new species of mammals not found 
in the "universal" paleontological record of the bituminous Pleistocene beds 



242 



of the Apsheron Peninsula. The occurrences noted in the Holocene of jackal, 
striped hyena, kulan and goitered gazelle and their present habitation of the 
area* indicate that their ranges in the Caucasus are relatively young. The 
southern origin and the migration of these species to the Caucasus in recent 
times can be deduced from the fact that the main areas of their distribution 
are to the south and east of the Caucasus, as well as from the fact that the 
species cannot survive deep and prolonged snow cover. In this species list, 
jungle cat, lion and tiger (so far unknown in the Caucasus as fossils) must 
probably be included. (This is discussed in more detail in Chapter III. ) 
246 Our paleontological materials indicate only one cycle of warming and 

development of xerothermal landscapes, which reached its maximum at a 
fairly early stage of the postglacial epoch. 

It seems possible that the fossil mammals of a country as highly 
diversified geomorphologically as the Caucasus may not reflect the minor 
climatic changes which might have occurred within the Holocene. 

The existence of a xerothermal stage was recognized by geobotanists 
Maleev (1939b, 1946), Yaroshenko (1956), Grossgeim (1948) and 
Takhtadzhyan (l94l). According to these authors this stage was responsible 
for isolated occurrences of xerophilous plants in western Transcaucasia 
and in some places in the Lessei- Caucasus uplands among more mesophytic 
areas . 

The xerothermal stage has also been mentioned by pedologist Yakovlev 
(1914) and Zakharov (1935), geomorphologist Shchukin (1925), zoologist 
Sviridenko (1927, 1937) and geologists Gerasimov and Markov (1939). 

The changes in the landscape caused by man in the last stage of the 
historical epoch are also traceable in the osteological material. 

The abundance of boar, deer, elk and roe deer in the food remains in the 
Sarkel beds of the Middle Ages indicates that extensive forests were 
developed in the Don River valley. By the 18th century, these forests had 
been exterminated and the result was the disruption of the ranges of 
distribution of some European species in the Caucasus. 

Occurrences of semifossil deer in the Stavropol area and the occurrences 
of boar and deer in some sections of the Armenian mountain -steppe (now 
arid) may to some extent provide confirmation of far fewer wooded areas 
in that country in Upper Holocene time. 

Studies of coals from the cultural beds of settlements of the second 
millennium B.C. (Tsalka) and the 12th century A. D. (Gandzha) have proved 
the existence of extensive forests in the present woodless areas of the 
uplands of western Georgia near the villages of Dabi and Gomi, in southern 
Georgia near Tsalka, and in Azerbaidzhan on the barren ridges of Kirovabad 
(Yatsenko-Khmelevskii and Kandelaki, 1940, 1941a, b). 

Our studies of the ritual collections of skulls in the mountain valleys of 
North Ossetia also indicated, through the disappearance of skulls of elks 
and bisons, that the forests on the sloping piedmont plains of eastern 
Ciscaucasia were exterminated by the 18th century. 

The paleontological materials studied provide many other data on changes 
in the ranges and dispersion of synanthropic species and on extinction of animal 
game through direct extermination by man and through the indirect effects 
of radical anthropogenic changes in the landscapes. 



With the exception of the extinct kulan. 



243 



247 Part Two 

ANALYSIS OF THE ORIGIN OF CAUCASIAN 
QUATERNARY MAMMALS IN RELATION 
TO THEIR DISTRIBUTION . ECOLOGY, 
AND MORPHOGENESIS 



249 Chapter HI 

REVIEW OF THE EARLY AND CONTEMPORARY 
POPULATIONS AND RANGES OF SOME MAMMALS 

The patterns of zoocenose formation can be traced in any given country 
most easily by an investigation of range development, with consideration of 
the ecology of separate species. Well-defined ranges form the basis of 
zoogeographical division. 

The existing range of a Palaearctic species may be only a relict of an 
ancient range, e.g., from the Quaternary period. 

Therefore, in order to reach a conclusion on species dispersal, its 
ecology must be considered against a background of paleographic and 
paleontologic data, current zoogeographical conditions and man's influence 
during historical times. 

The various factors which deternnine species survival or extinction 
divide into those of major influence and those of lesser influence. 

The investigations of the Binagady asphalt deposit (Chapter II) have shown 
that peculiarities of early and recent ranges of eastern Transcaucasian 
mammals depend primarily on the evolution of the Quaternary landscape. 
To a lesser degree these peculiarities are also dependent on morphological 
evolution and biocenotic interrelationships. It is thought that the 
phytophagous mammals behave autonomously during the development and 
extinction of ranges. The influence of predators on Herbivora is probably 
of secondary importance to their distribution. 

Theoretically the explanation of the origin and development of the range 
is more easily accessible from observation of colonial species — 
Carnivora, Proboscidea, Rodentia, Ungulata — remains of which are 
frequently found in various deposits. Ideally, these species should be 
comparatively stenotopic, that is, associated with a certain type of topography 
having a small nunaber of biotopes. Species thus widely distributed leave 
more fossil material and more recent evidence of early ranges than 
narrowly distributed species endemic to their relict ranges. Eurytopic 
animals yield excellent paleontological data, particularly if the investigator 
has a consecutive chronological series of deposits to study. 
250 The origin of other species, the fossil remains of which are rare, 

particularly the Insectivora and Chiroptera, may only be proved by means 
of zoogeographical mapping, analysis of present-day ranges, ecological 
analogy and extrapolations on the basis of more fully investigated specimens 
of Carnivora, Rodentia and Ungulata. 

An attempt has been made to trace the early and recent distribution of 
both living and extinct Caucasian mammals and the factors pertaining to 
their range and numbers. An investigation has also been made to determine 



245 



when these species appeared in the Caucasus and the routes taken by 
immigrant species. The examples selected include various orders and 
species. 



Order INSECTIVORA 

The paleontological age of generic series belonging to this order is quite 
ancient. The Insectivora fossils are known from the Eocene. Radial 
adaptation and the diversity of its relationships to the landscape are complex 
in the Insectivora. The earliest aquatic and widely specialized forms 
belong to 3 genera: D e s m a n a (desman), T a 1 p a (mole) and Crocidura 
(shrew) which were found in the Lower Pliocene complex of the Stavropol 
area (see Chapter II). Remains of 5 species belonging to 5 genera — 
Erinaceus, Hemiechinus, Crocidura, Neonnys and T a 1 p a — 
are known from the Pleistocene deposits. 

The contemporary fauna of the Caucasian Isthmus includes 15 species 
of Insectivora belonging to 8 genera and ecologically to three categories: 
fossorial, 1 genus and 3 species (mole); amphibious, 2 genera and 
2 species (desman and water shrew); and the remainder which are more or less 
specialized and associated with various biotopes (e.g., hedgehog, shrew). 



Family ERINACEIDAE 

European hedgehog — E r i пае eu s europaeus (s. lato). Fossil 
remains of Erinaceus species are known in Europe from the Miocene. 
Remains of the European hedgehog have been found in the Caucasus in the Middle 
Pleistocene strata of the Apsheron Peninsula and in Upper Pleistocene strata of 
western Transcaucasia. The hedgehog is widely distributed today (Map l), 
found from the Caspian Sea level to an altitude of 2,500 m throughout the isthmus. 
The hedgehog is absent from alpine meadows, semisteppe, semidesert, open 
desert and permanent bogs and is rarely seen in upland forests; it is 
most frequently observed in the lowland forests of Transcaucasia and the 
sparse forests and shrubbery of the rolling Ciscaucasian plains. 

This hedgehog lives in tugai thickets, gardens and oases of the semi- 
deserts in eastern Ciscaucasia and Transcaucasia. The range diminished 
in the Pleistocene over large areas of eastern Transcaucasia; the remaining 
251 ranges in the Apsheron Peninsula are attributed to the postglacial dryness 
of the area (according to Vereshchagin, 1949c). 

The origin of European hedgehog on the Caucasian Isthmus remains 
uncertain. Its range extends far to the north and south to the Caucasus. The 
absence of the hedgehog from open desert, semidesert and lowland steppes 
indicates its mesophilous character, its ancient associations with forest 
and its probable descent from Pliocene ancestors living in mesophytic 
environs of the eastern Mediterranean. Hedgehog remains from Middle 
Pleistocene deposits in Transcaucasia also confirm its presence in the 
Lower Quaternary, or more extactly pre-Quaternary, and the early 
distribution of this species in the Caucasus. 

Today the European hedgehog is spreading rapidly over the plain 
concomitantly with the development of steppe and desert irrigation. 

246 



Long-eared hedgehog — Н e m i e с h i nu s auritus Gmel. Remains of 
this species were found in Middle Pleistocene strata of the Apsheron 
Peninsula, and in Upper Pleistocene deposits of the northern Caspian 
lowland (Vereshchagin and Gromov, 1952). Hedgehog remains are known 
in the southwest from Paleolithic layers of Palestinian caves. Bones of 
the long-eared hedgehog carried into the caves by eagle owls have been 
found by the author in Holocene deposits in many areas of Ciscaucasia 
and Transcaucasia. 

Its contemporary range includes the dry lowlands of Ciscaucasia and of 
eastern Transcaucasia and the high plateaus of the south (Map 1). 

In the foothills of the Greater Caucasus, the hedgehog is found in the area 
between sea level and an altitude of 300-350 m. In southern Transcaucasia 
and Talysh it lives in the mountain-steppe and semidesert to altitudes of 
1,600- 1,800 m. The animal is absent from the mountain and lower forest 
zone. The largest population of the species is observed in the semidesert 
zone, i.e., the Kura-Araks lowlands. 

The adaptation of the long-eared hedgehog to a semidesert and steppe 
environment is expressed in its food specialization: mollusks, locusts and 
beetles, and, to a lesser extent, small reptiles and rodents. Low water 
requirements, the ability to endure high temperatures and its use of turtle, 
gerbil, suslik, fox and badger burrows associate this hedgehog with a 
southern xerophilous fauna. 

Extensive vertical distribution of the hedgehog in southern Transcaucasia 
and the relocation of the range southward indicates the probable penetration 
of the species from southwest Asia during the Upper Pliocene. 

The present range of the hedgehog is increasing somewhat with the aridity 
of the foothills, which has been brought about by man. 



Family TALPIDAE 

Mole— T alp a caucasica Sat., T. orientalis Ogn. The 
remains of the genus Talpa are known in Europe and North America 
from the Miocene (Trouessart, 1898-1899a; Simpson, 1945). The earliest 
remains of small moles (Talpa sp. ) in the Caucasus were in Pliocene 
strata near Stavropol. 

Remains of the recent species T. aff. caucasica are found in 
Pleistocene strata of western Transcaucasia. 

In Holocene sites they are very common, especially in the Ciscaucasian 
lowlands. 
252 Considering the early isolation of Caucasian moles from the European 

mole (Talpa europaea L. ) and the Mediterranean mole (Talpa coeca 
Savi) (cf. map by Bobrinskii, Kuznetsov and Kuzyakin, 1944), we class them 
as separate species (following Ognev, 1926a, 1928). The morphological 
adaptations of both species are similar; their sympatry is shown in 
Map 2. A recent study of species' ranges shows that Ciscaucasia, the 
Greater Caucasus and the Lesser Caucasus are inhabited by the Caucasian 
mole, with the subspecies T. caucasica ognevi Str. on the Rachin and 
Trialet ridges, whereas the eastern mole is found living in the Colchis 
lowland and the Talysh area. 



247 



The present distribution of the Caucasian mole is significantly related 
to the country's topographical development, as can be seen from the following 
features of its ranges. 

In western Ciscaucasia, the animal inhabits the Don, Yeya, Chelbas, 
Yegorlyk and Kuban river valleys whereas in the Kuban Plain the population 
declines during arid years and survives only in more humid depressions. 

There is an insular distribution of a large and varied population in the 
meadows and forests of the Stavropol Plateau. 

In the northern Caucasus, the mole is widely distributed in the foothills, 
but inhabits only the river valleys of the Taman Peninsula and Kabarda 
Plain, and is totally absent from the steppelike plains. The species is 
found in oak forests of Pyatigor'e and in the Podkumok valley as far as 
Georgievsk, and to the north along the Kuma River to Budennovsk, 
becoming rare east of Ordzhonikidze . 

The mole is common in upland forests of the northern slopes of the 
Greater Caucasus from the upper reaches of the Afips Basin to the Argun, 
especially along the humid Chernye Gory. It has been observed on the 
southern slopes of the Greater Caucasus from Gelendzhik to Zakataly and 
inhabits slopes covered by fir— maple forest in the Tsebelda, Inguri and 
Tskhenis-Tskhali ravines. 

In western Transcaucasia, the species inhabits the terraces of the Black 
Sea coast, along riverside ridges in swampy Colchis, and to the west on 
the northern slopes of the eastern Taurus Mountains. 

The mole is found at medium altitudes on the Surami ridge and in South 
Ossetia but not in the dry Gori depression. 

In the Lesser Caucasus, the range is widespread in the zone of broadleaf 
forests and subalpine meadows on the northern slopes of the Adzhar- 
Imeretian and Trialet ridges in Dzhavakhetia, and on the Bambakskii and 
Bezobdalskii ridges. 

Although the mole occurs in forests at altitudes of 400- 1,700 m inTalysh, 
it is rare because of the marked aridity in summer; (In July 1945, the 
Vilyazh-Chai, Vassaru-Chai and Vasharu-Chai ravines yielded only 3-4 moles 
per day from 100 traps. ) 

It is probable that the mole lives on the northern slopes of the Elburz 
Range and it survives in certain parts of the northern slope of the Sevan 
Range in the upper reaches of the Dzegam and Shamkhor rivers. 

Unexpectedly, it also survives in the semidesert zone of the Karayazy 
Plateau east of Tiflis, from a more humid epoch (Vereshchagin, 1940a). 
Here it lives in oak forests and motley-grass meadows along the banks of 
the spring-fed rivulets of the Kara-Su River. (Near this area steppe vole 
and red-tailed gerbils inhabit the solonchak meadows of sea lavender 
and arboreal saltwort (Figure 106); this is a remarkable example of the 
mosaic structure of biotopes and coexistence of hygrophilous and xerophilous 
animals . ) 

At the upper reaches of the Akstafa-Chai and Debed-Chai, Dal' (1944a) 
observed the maximum density — 42 moles per hectare — in a mixed beech — 
hornbeam forest at an altitude of 1,770-1,800 m. The author observed moles 
on the Karabakh Upland in the forested Terter-Chai ravine at an altitude 
254 of 700-1,600 m, but they were relatively scarce. 

The mole is not found in the dry longitudinal valleys fromEl'brus to 
Dagestan or in the interior of Dagestan. Its absence from the eastern part 



248 



of the Main Range is principally attributed to the steepness of the slopes, 
the thin layer of soil covering the bedrocks of the upland and the marked 
aridity in the lower part of the beech— hornbeam forest zone which results 
in an irregular food supply. 

The Alazan-Agrichai lowland, which is humid in winter but arid in 
summer, is also characterized by an absence of mole, as are the environs 
of Sevan and the Daralagez ridge. 

There are no data on the presence of the mole on the Iranian Plateau and 
in eastern Anatolia. Blanford (1876) did not mention moles. Radugin (1917b) 
included the mole in his list of animals at the Khoi oasis (an altitude of 
approximately 2,000 m), probably mistaking the mounds of the mole vole 
for mole hills . 
(252) Loginov (1949) obtained the following data on the relative density of the 
mole population from animals caught in July 1937 in various vertical zones 
(Table 64). 



TABLE 64. Comparative population of the mole in different 
environments of the northwestern Caucasus 



Zones 


Number of moles 
caught per day 
with 100 traps 


Forest-steppe of foothills 


40 


Broadleaf forests 


30 


Dark coniferous forests 


5 


Alpine meadows 


10 







253) 




FIGURE 106. Habitats of the mole (in oak forests) and red-tailed gerbil (on meadows with sea lavender 
and saltwort ) in the Karayazy forest of the Kura valley 

Photograph by author, 1958 



249 



This survey of the collective range of Caucasian moles indicates that 
their distribution is more closely associated with a landscape of broadleaf 
forests and mesophilous meadows, and less so with a landscape of motley- 
grass steppes. The mole's existence depends on narrowly limited physical 
properties of the soil, particularly humidity and food reserves, i.e., a 
considerable biomass of invertebrates. Migrations of the mole through dry 
steppe and semidesert are improbable; it is more likely that the now 
separated ranges of the Greater Caucasus, Lesser Caucasus and Talysh 
were formerly connected. The collective range of Caucasian moles indicates 
that the areas of their distribution shared a uniform landscape only in the 
Tertiary. The Caucasian mole range (most of which lies within the western 
regions) was already established by the Upper Pliocene following the early 
development of arid zones in eastern Ciscaucasia and Transcaucasia. The 
present discontinuity of mole ranges in the Caucasian Isthmus is a result 
of the "struggle" between mesophytic and xerophytic environments during 
continuous climatic changes and of orographic peculiarities in the Pliocene 
and Pleistocene. 

The mole migrations from the Russian Plain to the Caucasus and back 
obviously took place in the Neogene and Upper Pleistocene through the 
Trans-Kuban Plain and the Stavropol area. However, the Caucasus was 
also an independent focus of mole species formation. 

The ranges in Ciscaucasia, on the Armenian Highland and in the Kura 
River valley near Karayazy are probably of postglacial origin. The broken 
range in Talysh-Karabakh may be of an earlier, pre-Quaternary age. 

The contemporary range of the mole in the Caucasus has been reduced 
by forest felling and the xerophytization of the foothills and mountain regions 
during grazing and plowing. Only in the western Caucasian foothills is the 
direct effect of trappers on moles noticeable (Graph l). Pelt trade does 
not reflect variations of population. 

Desman — Desmana moschata L. The desman bones recorded from 
Pleistocene strata of southern and central Europe have been described 
(Trouessart, 1898-1899a; Wolf, 1938, 1939) as subspecies of the modern 
D. moschata hungarica Kormos, D. moschata fossilis Lartet 
and D. moschata magna Owen. 

In the Caucasus remains of Desmana sp. (mandibular fragment in 
gravel) were found in the ancient alluvium of a Pliocene river in the vicinity 
of Stavropol. Pleistocene remains are known from Paleolithic settlements 
in the Ukraine and fronn Quaternary alluvium of the Dnieper, Don and 
Volga rivers. 

The present range of the desman includes the basins of the Don, Volga 
and Ural. Only a small part of this area along the Don valley, from the 
255 Manych to the Don estuary where the animal has always been rare, is today 
adjacent to the Caucasian Isthmus (Map 2). 

The pronounced morphological and physiological adaptations of the 
desman to an amphibious existence in floodplain lakes supplied with 
Invertebrates indicates its ancestral evolution in river basins with a very 
stable hydrologic regime peculiar to rivers in extensive plains. 

The desman probably could not exist in mountain rivers of the Pliocene 
and Pleistocene Caucasus because of heavy floods and continuous erosion. 
Therefore we may consider the Eastern European species as a rather early. 
Upper Pliocene settler, but only in western Ciscaucasia near the Yeya and 
Chelbas rivers, probable tributaries of the paleo-Don. 



250 



256 



In fact some floodplain lakes are suitable for the desman, the lower 
Kuban, for instance, near the Cossack village of Grivenskaya and in the 
basins of the Colchis rivers Tsiya, Tsiva and Pichera. 



Order CARNIVORA 

The origin and age of the range area of most Caucasian Carnivora can 
be determined more or less from paleontologic data and from the 
peculiarities of the range. The small number of species and the 
rarity of mass destruction did not provide conditions for the formation of 
natural deposits of Carnivora. However, remains of predators are found 
in kitchen middens of Paleolithic man and in later settlements where these 
animals were hunted for their pelts. Gromov (1948) submitted a valuable 
stratigraphic analysis of the remains of Quaternary Carnivora. Fossil 
Canidae, Ursidae, Hyaenidae, Mustelidae and Felidae are known from the 
Cenozoic deposits of the Caucasian Isthmus. The species of these families 
form the present fauna of Caucasian Carnivora. 



Family CANIDAE 

From Pleistocene fauna of the Caucasus 5 species of Canidae of the 
genera Canis and Vulpes are known; of these two are becoming extinct. 
There are 5 existing species of the Holocene fauna. 

Jackal — С a n i s aureus L. The remains of the jackal are often found 
in fossiliferous cave deposits in southern Europe (France, Italy, Rumania), 
north Africa and southwest Asia. 

The difficulty of analyzing small Canidae according to their skeletons 
throws any identification of small Pleistocene European dogs with 
contemporary African and Asian jackals in doubt, especially since remains 
of Pleistocene jackals have not been found in Eastern Europe and Soviet 
Central Asia. Fossils in Acheulean and Mousterian cave deposits in 
Palestine, Lebanon and Syria may be considered closely related to the 
present species (Bate, 1937). In Pleistocene strata of the Caucasus, jackal 
remains have not been found, indicating that the jackal appeared on the 
Isthmus recently (Vereshchagin, 1949c, 1951b). Later remains of jackals 
were found in early Bronze Age settlements near Anaklia, in caves near 
Kutaisi and in the vicinity of the Saraibulakh ridge in Armenia (Dal', 1940b). 

The contemporary jackal is mainly restricted to the plains (Map 18). 
In western Ciscaucasia the jackal is rare; in eastern Ciscaucasia it 
inhabits the valleys of the Terek and Sunzha from Mozdok and Grozny. On 
the Black Sea coast the jackal is numerous from Gelendzhik to Batumi, but 
is not found at altitudes higher than 400-500 m. The jackal is particularly 
numerous in alder thickets of the lower reaches of Colchis rivers. Eastward 
the range passes through the Surami which only a few animals traverse. 
The largest population in eastern Transcaucasia is observed on the plains 
of Zakataly-Ismailly, Khachmas and Lenkoran. In winter jackals are 
numerous intugai thickets and margins of reedgrass bogs of the middle 
Kura and lower Araks. In fall when jackals feed on fruit, in particular 



251 



medlar, they are found at altitudes to 1,000 m in the Greater Caucasus. 
They also inhabit the forest zone of Talysh to its upper edge at 1,700- 
1,800 m. 

In summer the jackal lives in the open semidesert while feeding on 
locusts but winters there only if small hibernating bustards are plentiful. 

The high arid plateaus of northern Iran and Armenia are not inhabited 
by the jackal; it is also rare in the deep ravines. 

Despite its apparent adaptation to a moderate climate, the jackal is a 
thermophilous predator, unable to endure deep snow. Its southern origin 
becomes apparent during strong snowfalls in the Kura Lowland, when the 
animal becomes helpless and unable to move or to procure food, and often 
perishes from exhaustion and cold (Vereshchagin and Dyunin, 1949). The 
presence of the jackal among the Holocene fauna of southern Europe and its 
absence from the Russian Plain are probably explained by the difference 
in the thermal and snow regimes. 

The jackal is now vigorously hunted everywhere, but the population 
remains large. In the 1930's, 113 jackal pelts were obtained per 
1,000 km in Azerbaidzhan, 32.4 in Georgia and 10 in Dagestan 
(Vereshchagin, 1947). In the last 25 years there has been a considerable 
increase in the pelt yield over that recorded in 1934 (Graph 2). In 1949 the 
pelt yield increased 2.5 times because of the higher premium placed on the 
jackal as a menace to cattle and an unbalancing factor in the economy. This 
fact demonstrates that the jackal thrives in the existing environment of 
Transcaucasia. 

Wolf — Canis lupus L. (s. lato). Small wolves belonging to 
the polymorphic group C. lupus existed in the Upper Pliocene 
and Lower Pleistocene of Europe. Remains of a large wolf (C. lupus 
have been found in strata of the period of maximum glaciation. 
Those found in Germany and Austria are especially characteristic 
of the Wurm age (Freudenberg, 1914). 

In the Caucasus the most ancient evidence, a mandible belonging 
to a small wolf (C. tamanensis), comes from Lower Pleistocene 
conglomerates of the Taman Peninsula. The relationship of this species 
to contemporary C. lupus is uncertain. 

Remains of large wolf appear in this locality from the Lower Pleistocene 
on, and are common even now (Map 19). 

South of the Caucasus, C. cf. lupus remains are found in Mousterian 
layers of Palestinian caves (Bate, 1937) and in Upper Pleistocene talus of 
northwestern Iran. 

In the north the remains of large Middle and Upper Pleistocene wolves 
are common on the banks of the Dnieper, Don, Volga and Ural and in 
Paleolithic settlements of the Russian Plain and the Crimea. 

The present-day distribution of the wolf oh the Isthmus is general 
although not uniform. 
257 According to State figures of pelt yields, the greatest number of wolves 
obtained per land-unit area is from western Ciscaucasia and eastern 
Transcaucasia (Map 19). In the 1930's the yield of wolf pelts per 1,000 km^ 
was as follows: Azerbaidzhan — ll.fi, Armenia — 9.9, northern Caucasus — 
6.7. Dagestan — 6.7, Georgia — 0.8. 

Most pelts delivered to the State depots came from the plains and were 
taken from animals killed in winter. 



1704 



252 



The lower reaches of the Terek, Sulak and Kuma rivers and the Kura 
lowland coast of the Kyzyl-Agach Gulf are densely populated by wolves which 
feed on domestic cattle, wild boar and fish. 

In the alder bogs of Colchis the wolf is common in the lower reaches of 
the Gagida, Okum and Pichera rivers. 

In the mountain regions of the Caucasus a large wolf population can be 
observed in the Caucasian, Borzhomi and Zakataly-Lagodekhi reservations. 
The wolf populations settled in these areas, like those that live among the 
bulrushes of the Caspian coast, make predatory expeditions uphill and 
cross-country after wild ungulates. The wolf is distributed almost 
uniformly in all zones of the Caucasian Reservation, but in winter it is 
widely concentrated in the broadleaf forest zone. At this season large 
populations of boar, roe deer and deer abound. The zonal distribution of wolves 
found in the reservation (expressed in percentages) is as follows: 
alpine zone — 30%; dark coniferous forests — 32%; broadleaf forests — 
38% (Teplov, 1938b). According to Teplov, the loss of young boar, deer, 
roe, chamois and Caucasian buck killed in the reservation by wolves ranges 
from 34-61 % in afive -month period. In addition to the wild hoofed mammals, 
the wolf kills hundreds of domestic animals on lands adjacent to the 
reservation. 

The abundance of wolves in the Borzhomi area was noted by Dinnik 
(1914a). The wolf is endemic in the Zakataly Reservation as in the 
Caucasian Reservation (Markov and Mlokosevich, 1935; Vereshchagin, 
1938a). 

In other areas of the Isthmus, especially in the east, the wolf makes 
regular predations after herds of domestic animals as they graze — in the 
mountains during the summer and in the lowlands during the winter. In 
spring the greatest concentration of wolves is observed in arid foothills 
where there are guUeys and rock shelters; here the young are reared. 

The fluctuation of the wolf populations on the Isthmus during the Holocene 
can only be represented in a general outline. 

The number of wolves decreased in proportion to the decrease of wild 
Ungulata on the plain but gradually increased with the development of cattle 
breeding. The improvement of firearms and methods of capture and the 
breeding of large dogs for herd protection have probably reduced the 
number of wolves. 

The sharp increase of wolf pelts after World War II when the bounty 
was raised showed the stability of Caucasian populations of this animal 
(Graph 3). 

The Holarctic distribution of the wolf in the Quaternary, its poorly 
investigated geographical variability and the imperfect paleontological data 
leaves its center of origin unqualified. It is considered a Pliocene dweller 
of the Caucasus. 
258 Fox — Vulpes vu Ipe s L. (s. lato). The Holarctic genus Vulpes 
dates from the Upper Miocene. Remains of Pliocene foxes, V. meridi- 
onal is Nordm. and V. moravicus Wold., attainingthe size of the present 
central Russian fox, are known from Eastern Europe. 

Remains of Lower Pliocene foxes have not yet been recorded from the 
Odessa catacombs and the Upper Pliocene strata of the coast of the Sea of 
Azov. On the Caucasus remains of V. khomenkoi with a dentition 
resembling that of red and arctic foxes were found in fossiliferous strata 



253 



(Middle Pliocene) of the Apsheron Peninsula (Bogachev, 1938c, 1944; 
Vereshchagin, 1951b). An isolated fox premolar was found in Pliocene 
strata near Stavropol. 

The remains of foxes of the V. vulpe s group are commonly found in 
various types of Pleistocene and Holocene deposits of Western and Eastern 
Europe, the Crimea and the Caucasus, including Paleolithic encampment 
sites. 

South of the Caucasus remains of foxes have been found in Paleolithic 
strata of northern Iran (Coon, 1951), Syria, Lebanon and Palestine, from 
which Picard (1937) recorded V. alopex for the Mousterian and 
Aurignacian, V. cf. niloticus Rupp. for the Acheulean and Mousterian, 
and Canis vulpes for the Aurignacian and Mesolithic. 

The present-day distribution of the fox in the Caucasus is general 
(Map 20). The greatest density of the fox population is observed in the 
zones of semidesert, steppe, foremontane forest steppe and highland steppe; 
the density is least in alpine meadows. Foxes are rare in the mountainous 
beech forests of the southern slopes of the Greater Caucasus. Dinnik (1914a) 
refers to the thinned forests on the foothills and plateaus of the northern 
Caucasus when discussing the fox population in Caucasian forests. In the 
CaucasianReservationthe largest number of foxes, upto43.1%, inhabits the 
broadleaf forests throughout the year; a smaller number, up to 31.8%, 
is found in the spruce— fir forest zone, and the smallest number, up to 
25.1%, in the alpine and subalpine zone Khonyakina, 1938). This distribution 
generally corresponds with the particular food supply: Muridae, birds, 
insects and berries. A different situation exists in Transcaucasia. 

In 1935-1936, we found indications of fox distribution on the southern 
slopes of the Greater Caucasus in Azerbaidzhan in the form of fresh foot- 
prints and feces. The distribution according to the vertical ecological zones 
in which the foxes were found is given in Table 65. 



(259) 



TABLE 65. Frequency of foxes found and their tracks in different zones of eastern Transcaucasia 





Number of foxes and their tracks 
recorded on a 20 km walk 


Land types in eastern Transcaucasia 


1935 


1936 




June 


January 


June 


January 


Virgin semidesert 

Lower forests of the Agri-Chai valley 

Beech — ^hornbeam mountain forests 


2 


4 

1 

1 


8 

4 
1 
1 


12 
6 
2 


Alpine meadows 


2 


Rocks and taluses of passes 









In the semidesert zone the slopes of gulleys and foothills are densely 
populated in spring, whereas the seashore and foothills are populated in 
winter. 

The basic food in summer consists of grasshoppers, and in winter of 
beetle larvae, caterpillars, steppe vole and gerbils. The seashore attracts 



254 



the foxes because of its abundant food supply: dead birds, fish, crayfish. 
The contemporary fox distribution according to biotopes is greatly affected 
by man's agricultural activity (cf. Chapter VI). 

The yield of fox pelts during 1930-1940 showed the densest population 
of foxes inhabiting the plains of Ciscaucasia and eastern Transcaucasia. 
The following numbers of pelts were obtained per 1,000 km of territory: 
northern Caucasus — 276; Azerbaidzhan — 268; Armenia — 158; Dagestan — 
138; Georgia — 74. 

The number of foxes is inversely proportional to the degree of afforesta- 
tion and to the mountainous relief of the country (Vereshchagin, 1947). The 
fox is especially rare in western Transcaucasia which is rich in forests and 
mountains. 
259 According to Bakeev's records, the density of the fox population in the 
Spitsevka area of the Stavropol region during the period 1938-1941 was 12.5 
animals per 1,000 hectares in summer and 10.3-16.5 in winter (Chirkova, 
1952). 

During historical time the range and population of foxes in the Caucasus 
has apparently thrived under the favorable influence of agriculture, and 
even concentrated gathering has failed to reduce their nunnbers. 

The data on pelt supply during the last 25 years show sharp variations 
in the number of pelts for Ciscaucasia and fewer variations for 
Transcaucasia. Transcaucasia is more diverse ecologically and topographi- 
cally and therefore the fox population is more resistant. From the pelt 
yields of 1925-1952, three peaks in population increase can be noted — 1927, 
1936 and 1946 —with corresponding decreases in 1931, 1941 and 1951 
(Graphs 4 and 5). 

Considering the presence of similar factors in the Middle Pliocene of 
the Caucasus and the marked polymorphism of the present species in the 
Isthmus (Chapter IV), the fox should be included in the local eastern 
Mediterranean species of the Upper Tertiary. 

Corsac fox — Vulpes corsac L. The range of the corsac during the 
Pleistocene extended over a wide steppe belt from Western Europe to North 
China. Remains of this fox have been found in caves of Switzerland and 
Czechoslovakia. They are known also from a series of Paleolithic 
encampments in the Russian Plain and particularly from the Kodak and 
Crimean caves. 

On the Caucasian Isthmus corsac remains were found in large numbers 
but only in the Middle Pleistocene sands of Apsheron. To the south, 
particularly in Syria, Palestine and Iran, the corsac has not been observed. 
Here it was replaced by other mountain-desert forms of small foxes. By 
the 20th century the corsac was widespread in the Caucasus, and 
particularly in the northeastern part of the Isthmus (Map 21). 

The reduction of the western part of the species range probably began 
in the Upper Pleistocene but has only been apparent during the last 150 years 
because of the pelt trade. Since the Middle Pleistocene the corsac range on 
the west coast of the Caspian Sea has receded to the north for a distance 
of 700 km. 

Why the range diminished in Europe and the Caucasus since prehistoric 
times is still open to question. The data available indicate that the changes 
260 in group structure of small steppe predators occurred in the Middle 

Pleistocene. In Western Europe, for instance, a series of small wolflike 
and jackal-like Canidae became extinct without leaving a trace. 



255 



At the time of the Binagady asphalt trap the corsac population was still 
fairly large in the biocenosis of the semisteppe of eastern Transcaucasia, 
but was subsequently depleted. The number of corsacs in relation to fox 
and wolf (as calculated by the authors for the Quaternary from bone 
remains and pelt yields) is presented in Table 66. 



TABLE 66. Comparative numbers of Quaternary Canidae (in %) 



Date and locality 


Corsac 


Fox 


Wolf 


Holocene, 20th century. Mean annual pelt yield 
in the Northern Caucasus for the period 
1935-1940 — 63 600 


0.06 

6,8 
25.6 


97.5 

77.3 
37.7 


2 4 


Holocene, Sarkel, 8th- 13th centuries A. D. Total 
number of the specimens trapped by Khazars 
and Slavs — 44 

Middle Pleistocene, Binagady, Total number 
of specimens trapped in asphalt lakes — 327 


15.9 
36,7 



The corsac has long been a less vital form than fox or wolf, although in 
those regions most favorable for it the corsac population still exceeds those 
of fox and wolf even now. For example, south of Akmolinsk in the grass — 
wormwood steppe 37 corsac burrows, 3 fox burrows and 1 wolf burrow 
were found in 1947; in 1948, 42 burrows were observed, 34 of which 
belonged to the corsac and 8 to the fox (Chirkova, 1952, p. 402). 

The corsac apparently originated in the steppe region of Kazakhstan. 
It is a steppe-desert animal, feeding on small rodents, birds, reptiles and 
insects. 

The corsac was common in the steppes of the western Manych area and 
could still be found near Salsk between 1900-1910 (Dinnik, 1914a; Bogachev, 
1918). At present only a few pelts are obtained from this region (Graph 6). 
Steppe plowing and man's agricultural activity exerts an adverse influence 
on the corsac; now the animal is common only in the region of the Terek- 
Kuma sandy plains. 

The range of the corsac, considered from both its contemporary ecology 
and paleontological data, indicates that this species was a typical middle 
Pleistocene migrant from the Russian Plain to the Caucasus. The migration 
must have occurred during the maximum southward advance of the cold- 
steppe landscape. 



Family HYAENIDAE 



261 



Hyenas and Ichthytheriinae remains are known in Eurasia from the Upper 
Miocene (Simpson, 1945/. They are especially characteristic of 
Mediterranean Pliocene deposits. 

Three or four fossils and one living species of hyena are known from 
the Caucasus. Earlier remains were found in a Middle Miocene deposit near 
Belomechetskaya (Map 21). 



256 



Two specialized lines emerged in the Upper Miocene: Crocuta, with 
carnassial teeth resembling those of large cats, and Hyaena, more 
primitive and with less powerful teeth. 

In the Quaternary there were two species. Their exact relationship to 
the Tertiary forms is not clear. 

Cave hyena — Crocuta spelaea Goldf . This species is known from 
two localities in the Caucasian Isthmus where it apparently lived during the 
Pleistocene: the Middle Pleistocene strata of the Apsheron Peninsula 
(Binagady) and the Upper Pleistocene deposits of the Kuban Plain (ll'skaya). 

North of the Isthmus the distribution range of this hyena encompassed 
enormous areas of Eurasia in the Pleistocene (Pidoplichko, 1951). 

To the south the remains of this animal have been observed in the Upper 
Pleistocene of northwestern Iran (Brandt, 1870; Coon, 1951). 

The Middle Pleistocene Crocuta spelaea Goldf. of the Russian 
Plain was considerably larger than the Transcaucasian species 
(Vereshchagin, 1951b), and corresponds to the larger sizes of Herbivora 
of these regions, such as bison and horse. 

According to cranial features and body size, the Binagady hyena occupied 
an intermediate place between the Crocuta spelaea Goldf. and the 
extant C. crocuta L. of Africa. A direct relationship is quite possible. 

The distribution of С r о с ut a spelaea Goldf. decreased rapidly in 
the Upper Pleistocene, moving southwestward to Africa, but it is possible 
that a modified species inhabited an area south of the Isthnnus in the Holocene. 
Thus Crocuta crocuta L. inhabited Palestine in the Mousterian, 
in the Aurignacian and in the Holocene "Natufian" with the reappearance 
of the xerothermic climate (Bate, 1937). 

This animal's extinction in Eurasia is of particular interest; undoubtedly 
one factor was the gradual extinction of Proboscidea and large Ungulata in 
the Upper Pleistocene. Paleolithic man could scarcely have effected the 
disappearance of this hyena; even the present-day hunter with his modern 
arms has not wiped out the species. Pidoplichko (1951, p. 160) states that the 
anthropic factor is negligible in the extinction of hyena from Europe and 
Asia. The Mediterranean Sea area can be considered the native country 
of Crocuta spelaea Goldf. The expansion of its range should be 
investigated as well as the distribution of the mammoth fauna. The 
advance of the postglacial xerothermic period was marked in the Caucasus 
by the migration from the south of a less specialized species, the striped hyena. 

Striped hyena — Hyaena hyaena L. Fossils of the striped hyena 
were found in Paleolithic caves of Palestine, Morocco, Algeria and 
Portugal. The finds reported in France are doubtful (Wolf, 1938, p. 59). 
Pidoplichko (l95l), following Sharff (1918), assumes that the striped hyena 
arrived in Europe in the Pliocene from the southwest because land extended 
from Africa to the British Isles in the Pliocene. There is no proof to date 
that H. prisca Mars, de Serres and H. m on s p e s s ul an a Croiz. etJob. 
^^2 from the Pleistocene in southern France and H. ant i qua Lankest. from 
the Pliocene in England are ancestral to the contemporary species. 

Remains of the striped hyena have never been found on the Russian Plain 
or in Siberia. Only in Soviet Central Asia, south of Samarkand, in the 
Paleolithic strata of the Aman-Kutan cave have fragments of bones and teeth 
of this species been found (collection of the Zoological Institute of the 
Academy of Sciences, U.S.S.R.). 

The absence of striped hyena remains from Pleistocene deposits of the 
Russian Plain, the Crimea and the Caucasus, especially from the Binagady 

257 



complex, verifies that it penetrated the Isthmus comparatively late. Striped 
hyena fossils are found on the Caucasus but only in Holocene deposits of the 
Apsheron Peninsula. The present-day range of a number of subspecies of 
this hyena embraces Soviet Central and southwest Asia and Africa. The 
range areas have a characteristically xerothermic climate. The underfur of 
this hyena is only slightly developed, another indication of its southern 
origin. Its ranges on the Caucasus are principally in the hot semidesert 
of lowlands and intermontane valleys (Figure 107) in eastern and southern 
Transcaucasia. At the close of the last century hyena was still found in 
eastern Ciscaucasia, between Derbent and Makhachkala (Dinnik, 1914a; 
Satunin, 1915a). Occasionally animals were observed near Vladikavkaz 
(now Ordzhonikidze) (Bogdanov, 1873). In the years 1861 —1869 many hyenas 
were killed near Tbilisi, Yerevan and Nakhichevan (Radde, 1899). Prior to 
1930 the hyena was found in small numbers and only in the sparsely populated 
areas of western Azerbaidzhan and in the western areas of Georgia 
(Vereshchagin, 1942b). From 1930 to 1940 the yield was 26 hyenas; from 
1940 to 1950 only 5 or 6 animals were caught. Today a small number of 
263 hyenas can be found in the discontinuous desert ranges of the Kartalinia 
Plateau, near Udabno and Geredzhi. 

The typical habitats of the hyena in eastern Transcaucasia are the eroded 
clayey areas grown withthickets of juniper and pistachio. 

During the last fifty years the range and population of hyenas have decreased 
very rapidly, mainly because of a planned extermination campaign provoked by 
hyena attacks on children. 

(262) 




FIGURE 107. Habitat of striped hyenas in the Middle Araks vaUey (Abrakunis) 



Photograph by author, 1947 



Family URSIDAE 

From the Cenozoic strata of the Caucasus two representatives of Ursidae 
are known: Ursavus and Ursus. 



2S8 



Ursavus is a Middle Miocene genus found in Belomechetskaya. 
и r s u s is a lower Pliocene genus known from a Pliocene deposit near 
Stavropol in the Caucasus (Maps 22 and 23). 

The Quaternary fauna of the Caucasian Isthmus includes three species 
of bear, genus U r s u s , of which only one lives today. 

Ursus (Spe lae arctos ) rossicus Boris, was reported in 1931 
from five skeletons recovered from sandy deposits of ancient alluvial flows 
(Middle Pleistocene) near Krasnodar (Figure 108). 




i i I i I I I I Г' > 

FIGURE 108. Skull of Ursus (Spelaearctos) r ossi с us Boris. , from Krasnodar 



The remains of Ursus (Spelaearctos) rossicus Boris . were also 
observed near Kherson (Borisyak, 1931), in the basins of the Greater Irgiz 
(Belyaeva, 193 5, 1939a) and the Ural rivers (Vereshchagin and Gromov, 
1952) and are now reported to be in caves of the northern Urals. Since 
remains of this species do not occur in Paleolithic strata of the caves of 
western and central Transcaucasia and in the bituminous strata of the 
Apsheron Peninsula, we can assume that the southern limit of its range 
was somewhere in the Ciscaucasian foothills. Ursus (Spelaearctos) 
rossicus Boris . probably lived in river valleys and steppe ravines and on 
slopes of foothills and was not as closely associated with karst topography as 
Ursus (Spelaearctos) spelaeus. The broad molar cusps indicate 
the preeminently phytophagous nature of the animal. 
264 We consider this bear to be a Lower Pleistocene settler in the Russian 

Plain; it probably became extinct during the Upper Pleistocene. 

Cave bear — Ur sus (Spelaearctos) spelaeus Rosenm. During 
the second half of the Pleistocene the cave bear was a species associated 
with the landscape of Western Europe and north Africa. 

It is known from the remains of the Il'skaya camp site near Krasnodar 
in the Caucasus and from eastern Transcaucasian caves (Figure 109). 
Reports on the presence of this bear in the Binagady complex by Bogachev 
(1939, 1940b) and Gromova (1948) are erroneous (Map 22). 



259 




FIGURE 109. Skull of Ursus (Spelaearctos) spelaeus Rosenm. with 
shortened facial part from Ablaskira cave 



The slight fossilization of the remains of the cave bear from the 
Vorontsovskaya cave near Khosta is very interesting. It is possible that a 
small cave bear population survived the Pleistocene in northeastern 
Transcaucasia. 

The remains of this species are unknown in the Russian Plain, except 
north of the Black Sea. Near the Samara Bend, however, a small cave 
bear existed in the Pleistocene which according to the skull and dentition 
was intermediate between the small and large cave bear. 

On the Lesser Caucasus, in Talysh and south of the Isthmus 
in Kurdistan and in Syria, cave bear remains have not been found. There- 
fore western Transcaucasia must be considered for now as the southern 
limit of the species range in the eastern Mediterranean. 

The affinity between this species and the small cave bear should be 
further investigated; there is no reason, however, to consider this Upper 
Pleistocene species ancestral to the small cave bear because fossils of 
both forms exist within close geological proximity. From its origin, the 
large cave bear can be classified as a local eastern Mediterranean species. 

European brown bear — и r s u s arctos L. (s. lato). In comparison 
with the cave bears the European brown bear represents a more progressive 
branch which at the same time retains a number of primitive features. No 
less than ten forms are recorded from the Pleistocene cave deposits of 
Western Europe, north Africa and southwest Asia (Syria). 

The European brown bear was widespread on the Caucasian Isthmus 
265 during the Upper Pleistocene; this is substantiated by fossils found in the 
caves of Imeretia and in Apsheron Peninsula talus. 

North of the Isthmus the remains of Ursus arctos L. (s. lato) were 
found in Pleistocene sands and pebbles in the river valleys of the Don, Volga 
and Kama, and in the diluvium of several Paleolithic encampments on the 
Russian Plain. Paleolithic U. cf. arctos and U. syriacus have been 
substantiated in Lebanon, Syria and Palestine (Picard, 1937). 

While remains of the European brown bear are rare in Holocene 
deposits of the Caucasus, the species is nevertheless found as a regular 
replacement of the large brown bear in multilayered deposits indicating its 
broad dispersal during the postglaciation (Map 23). 



260 



266 




FIGURE 110. 

1-2 — bronze pendant — bear figure (1:1) 
from the Kobanian burials of Ossetia 
(according to Uvarova, 1900); 3 — picture 
of a running bear(x 2) on a carnelian seal 
from Urartu graves (Orig. ) 



Pictures of bears are comparatively 
rare in the ancient art of the Caucasian 
people. In Kobanian burials of North 
Ossetia pendants and seals of bronze 
representing a bear were found. A 
running bear is shown on a carnelian 
seal (Figure 11 0) from an Urartu grave 
on the shore of Lake Sevan. A bear 
climbing a tree is figured on a silver 
container from the Maikop burial. 

The contemporary range of the bear 
in the Caucasus is still extensive. On 
the northern slope of the Greater 
Caucasus it extends from Novorossiisk 
in the west to the upper reaches of the 
Dzhengi-Chai in the east, wherever 
there are forests, mountain shrubs and 
meadows. The greatest population of 
this animal was observed in the 
Caucasian Reservation. On the Stavropol 
Plateau the bear was exterminated in the last century and is only rarely 
observed in the forests of Pyatigor'e. There are few bears in the Kabardian 
A. S. S. R. and in northern Ossetia, and none in the largest part of 
central Dagestan, but they are still seen in the Deshlagar forests. The bear 
was exterminated about 30 years ago in the lowland forests of Samur and 
on the gently sloped Kusary Plain. 

On the southern slope of the Greater Range there are now more bears 
than on the northern slope. Bears are common in wooded Abkhazia and 
Imeretia. There are no bears in the Colchis swamp. In Svanetia and South 
Ossetia the bear is common in reaches of the Kodor, Ingur and Rion, despite 
energetic attempts by the local population to exterminate the animal. In 
Azerbaidzhan it is numerous in the basins of the Belokan-Chai, Katekhi- 
Chai, Kakhi-Chai and other rivers, as far as the upper reaches of the 
Gerdyman-Chai in the east. 

In September and October when chestnuts and walnuts ripen, the bears 
concentrate in forests and gardens of the Alazan-Agrichai valley. The 
animals congregate on the upper forest edge when beechnuts and acorns are 
abundant in fall. On the low Kartalinia, lora and Adzhinour plateaus it lives 
in the sparse forests of juniper and Turkish terebinth where it feeds on 
terebinth berries, juniper and grasses, and especially on astragali. The 
bears track through gulleys and along precipices. Recently bears inhabited 
the tugai floodplain forests of the Kura, especially the section between 
Mingechaur and the Alazan estuary. 

At present the bear is uncommon on the Lesser Caucasus where it has 
been nearly exterminated. It is not found in woodless Dzhavakhetia. On the 
Armenian Highland it inhabits sparse juniper forests and relict oak forests 
and it is rare in mountain forests of Talysh. 

The greatest yield of recorded bear pelts was from eastern Trans- 
caucasia. From 1930 to 1940 the average yield per 1,000 km^ was as ' 
follows: Azerbaidzhan — 1 .5, Armenia — 1.5, Georgia — 0.9, northern 
Caucasus — 0.9 and Dagestan — 0.2. Most of the pelts, however, were 
retained by the local population. 



261 



The European brown bear of the Caucasus was probably of endemic 
Pliocene origin. This local origin is substantiated by Pliocene finds neai' 
Stavropol, by its abundance in the area and by the diversity of its forms. 
In different parts of the country, however, new genetic influences from the 
northern and southern neighboring populations were possible. The recent 
Holocene relationship of the range areas of boreal and Caucasian bears 
existed in the Don valley, the steppe ravines and the scarps of western 
Ciscaucasia; this may explain the resemblance of the northwest Caucasian 
bear to the European bear (Smirnov, 1916a). 

Man's activity, which wrought changes in the landscape, also affected 
the population and range of the bear. The pelt yield increased somewhat 
in 1930 then dropped and increased again in 1941 (Graph 8). 

The future of the bear on the Caucasus is not promising because of 
continual forest felling and hunting of bear as a trade animal and cattle- 
breeding pest. 



Family MUSTELIDAE 

Remains of Mustelidae emerge in Eurasia in Lower Oligocene strata. 
From the Pleistocene fauna of the Caucasus 6 species of mustelids are known, 
one of which probably became extinct in the Holocene. 

Glutton— Gulo aff. gulo L. Glutton remains have been found in many 
caves from the Upper Pleistocene of the southern and middle belts of 
Western Europe. In the U.S.S.R. glutton remains were found in Crimean 
and Ukrainian Paleolithic strata on the Kama banks and in the Altai and 
Urals caves. 

A fragment of glutton mandible was found in the Upper Paleolithic strata 
of the Gvardzhilas cave and in the Lower and Middle Paleolithic strata of 
the Kudaro I cave. This was the most southeastern habitat of the 
Pleistocene species. 

Smirnov (1923-1924) considered the glutton a mountain animal and 
explained its extinction on the Eurasian plains by the disappearance during 
postglaciation of compacted snow on which this animal could overtake average - 
size ungulates. 

Pidoplichko (1951) explained the withdrawal of the glutton range to 
northeastern Europe and its rapid extinction by "anthropic factors. " It is 
interesting that a similar decrease in range is recorded for the arctic fox, 
the reindeer and other arctic animals. In contrast, the range of the glutton 
shows a wide gap between its early Mediterranean focus and its northern part, 
added during the Pleistocene. The Caucasian focus of the species became 
extinct only slowly because food was abundant and mountain topography and 
therefore snow conditions were diverse. Many factors were responsible 
267 for the extinction of this animal in the Caucasus, the most important reason 
being decline in reproduction in the present-day phase of species evolution. 

The glutton is an autochthon of the mountain margins of the Mediterra- 
nean area; from this focus the animal dispersed northward and can be' 
observed today on this new range. 

Exactly when the glutton became extinct on the Caucasus is as yet not 
determined. It is, however, possible that it became extinct Lhere later 
than in southwest Europe, perhaps in the Bronze Age. 



262 



Stone marten — М a г t e s foina Erxl. The genus Martes is known 
in Eurasia from the Lower Pliocene. If we include the real martens in 
genus Mu stela, we can assume that this collective group developed in 
the Holarctic by the Upper Miocene. Mustela filholi Goudry and 
Putorius sp. were reported by Bogachev (1938c) from the Caucasus in 
the Pliocene deposits of the Armenian Highland. 

Fragments of Martes sp. were collected by archaeologists in the 
Middle and Upper Pleistocene strata of western Transcaucasian caves: 
Akhshtyrskaya and Sakazhia. On the F^ussian Plain we found jaws of 
Martes sp. in the Pleistocene sands of the lower Kama. This species 
is probably related to the pine marten. Real martens were not found in 
lower streams of Russian Plain rivers, but in the Crimea they were common 
in Paleolithic strata. From the Palestine Paleolithic, a M. palaeo- 
syriaca Nehr. similar to M. foina has been recorded. The marten 
of Urartu time from the Sevan shores was incorrectly recorded as 
M. latifrons Sat, 

The contemporary range of the stone marten is enormous: it spreads 
from the shores of Spain along the middle latitudinal belt as far as northern 
China and Manchuria. The Caucasian subspecies M. foina nehringi 
Sat. inhabits forested foothills, mountain forests and woodless plateaus 
(Map 24). On the Stavropol Plateau and on the undulating plains west of 
Stavropol it inhabits shrub-grown ravines and scarp lines of Sarmathian 
limestones. There is an annual yield of pelts from the middle reaches of 
theKumainthe area of Budennovsk, and the Greater Caucasian foothills, the 
anterior valleys and central Dagestan are also inhabited by the stone marten. 
It is widely distributed in the basins of the left tributaries of the Terek, 
the Sunzha and the Manas, and in the Gunib District. An excellent pelt yield 
was taken from the lower forests of Samur. We found this marten near 
Baku, near the base of the Apsheron Peninsula. 

The stone marten is absent from the fringing forests of the Kura-Araks 
lowland. 

In western and eastern Transcaucasia Martes foina Erxl. dwells 
primarily in the foothills and lower mountain forests; it is very numerous 
in the lower forest areas, e.g. , in the Alazan-Agrichai valley, where 
pelts are generally obtained. On the Armenian Highland and in Talysh the 
stone marten is widespread wherever there are small shrub thickets and 
forests. It lives in colonies in the upper mountain zone. However, 
even in desert intermontane valleys, e.g., on the middle Araks, the 
animal inhabits only large rocky placers and valleys, and ravines of 
isolated summits such as Ilyanlu-Dag, Darry-Dag and others. 

Judging from its ecological distribution the Caucasian stone marten is an 
early Pliocene dweller originating in the south. 

The activity of man, especially trading, affected it less than the following 
species. 
268 Pine marten — M a r t e s martes L. The remains of the pine marten 

are common in Pleistocene deposits of Western European caves (in England, 
France, Italy and Hungary). They are also recorded from Mousterian 
strata of Palestine (Bate, 1937). It is difficult to distinguish the bones 
of this marten from the bones of the stone marten which probably lives in 
some Paleolithic encampments of eastern Transcaucasia. 



263 



The present-day range of the Caucasian subspecies M. martes 
lorenzi Ogn. is very peculiar; it mainly inhabits full-grown beech forests 
on the northern and southern slopes of the Greater Caucasus, gradually 
disappearing in the east. On the Lesser Caucasus it inhabits the northern 
slopes of the Adzhar-Imeretia ridge, and on the Trialet ridge it is found 
from Borzhomi and Bakuriani to the Belyi Klyuch. To the west, the range 
extends from the northern slope of Taurus to Asia Minor (Map 25). East 
of Debed-Chai and Akstafa-Chai the marten is very rare if it exists at all. 
No marten pelts have been obtained from Karabakh forests during the last 
25 years. 

Satunin (1907b) recorded a skull of a young marten from a site located 
on the eastern shore of Lake Sevan as M. martes (?). His diagnosis was 
based upon delicate features and small size; this skull, which is now in the 
Georgian Museum, is in fact that of a stone marten. 

The range of Martes martes L. had not reached Lake Sevan by 
the middle of the first millennium B.C. From the seven marten skulls taken 
from the Urartu graves not one belonged to Martes martes L. 

The species is absent from Talysh despite the abundance of forests and of 
the fat dormouse, a very important food. Ellerman and Morrison-Scott 
(1951, p. 245) incorrectly claim the presence of the pine marten near 
Asterabad. 

The absence of the pine marten from the eastern part of the Lesser 
Caucasus and the Talysh Mountains may be attributed to the late appearance 
of the species in the Caucasus and the presence of an ancient woodless 
barrier in the Khram basin. The competition of the stone marten in the 
southeastern regions of the country should also be taken into consideration. 

The greatest number of pine martens is observed in the western 
Caucasus. Pelt yield per 1,000 km2 for both martens- is given by pelt 
supply stations for the 1930's: Georgia — 37.8, Armenia — 30.3, 
Azerbaidzhan — 19.4, Dagestan — 10.3 and northern Caucasus — 7.8. 

The high number of martens in Georgia is explained by the heavy 
afforestation of the country. In Armenia the entire pelt yield was from 
stone martens. 

Donaurov (1949) observed the pine marten in the Caucasian Reservation 
between 1935 and 1936. (Its summer excrements and winter snow tracks 
were found distributed according to Table 67. ) 

Such a distribution is the result of an abundant supply of vegetable and 
animal foods: during summer months — Muridae and insects; in fall and 
winter —mountain ash, whortleberry, common yew and rodents. 

The small population of pine marten in the lower forest is probably a 
result of its replacement by the stronger stone marten. On the southern slopes 
269 of the Greater Caucasus the distribution of the animal is somewhat different. 

Table 68 shows the frequency of encounter of the marten activity in the 
Zaketaly Reserve and adjacent lands, according to our data of 1935—36. 

Our data and those of Donaurov for the lower and middle zones also 
pertain to the stone marten. 

According to Mlokosevich's observations (Markov and Mlokosevich, 
1935) made over several y^ars in the Lagodekhi and Zakataly regions, the 
pine marten inhabits the upper mountain forests and is unable to compete 
successfully with the stronger stone marten of the foothills. It is absent 
from the woodless plains. 
* The State Yield Supply recorded both marten species together. 



264 



TABLE 67. Frequency of pine marten tracks in the western Caucasus (in °/o) 



Habitat 


Summer 


Winter 


Broadleaf forests 


0.7 
16.9 

37,8 

40.9 

3.7 


6.9 


Fir— beech forests without vacciniaceous plants . . 
Fir— beech forests with underbrush and 
vacciniaceous plants 


9.4 
51.8 


Upper forest edge 


30.9 


Subalpine meadows 


1.0 







TABLE 68. Frequency of pine marten tracks on the southern slopes of the Greater Caucasus (in %) 



Habitat 


Summer 


Winter 


Beech— hornbeam forest of the lower zone .... 
Beech forests with mazzard cherry and filbert 
of the middle zone 


12.5 

30.6 

50.0 

7.9 


13.2 
47.8 


Thinned oak— beech forests of the upper zone . . . 
Subalpine meadows 


39.0 







The pine marten is a more specialized and phylogenetically younger 
species than the stone marten as shown by its morphology and ecology. 
The separation of pine marten probably occurred in Europe in the Lower 
Pleistocene since it was a distinct species by the Middle Pleistocene. The 
ecology of the pine marten's contemporary range confirms the relatively 
late (Upper Pleistocene) migration from Europe through the Balkans and 
Asia Minor. 

The direct and indirect influence of man on the population and range of 
Caucasian martens became visible in the last century. 

Ancient Circassia produced a heavy yield of pine martens. There are 
references in Dubois de Montpereux (1843) of the yield of pelts in this area 
in the 18th century and Radde (1866) records an annual yield of 500 marten 
pelts from upper Svanetia in the middle of the last century. Marten-hunting 
for pelt-trading was taken up in the 1860's following the settlement of the 
l270 northern Caucasus (Dinnik, 1941a; Turov, 1937). The clearing of significant 
areas of forest land over the past centuries in Dagestan and Azerbaidzhan 
could have brought about a reduction in the marten range. Records of the 
last 25 years show a decrease in pelt yield in 1930-1931, an increase in 
1937 and another in 1947. Some drop in trade in Abkhazia was observed 
in the 1930's as an apparent result of the development of the fat dormouse 
trade (Markov, 1938; Graph 9). 



265 



Family FEUDAE 

At least 10 forms of cats have been identified from Neogene deposits 
of the Caucasus and adjacent areas. Machairodus sp. is known from 
Sarmatian deposits near Armavir and a medium-sized Felis sp. from the 
Maraginsk Upper Pleistocene site in northwestern Iran. More fragmentary 
bone findings of Felis and Panthera have been reported from the 
Lower and Upper Pliocene of Stavropol and the Taman Peninsula. 

Six species have been recorded from Caucasian fauna of the Pleistocene: 
Panthera spelaea Goldf . , panther, lynx, European wildcat, spotted 
cat and cheetah. And finally from the Holocene there have been nine 
species identifications; of these nine, three have become extinct in historical 
time. 

Cave lion — P an t h e r a spelaea Goldf. Remains of this species 
of large cat have been discovered in Middle and Upper Pleistocene 
Eurasian strata — from England to the New Siberian Islands (Trouessart, 
1904-1905). It is a characteristic index species of the Pleistocene mammoth 
fauna of the Russian Plain and of the Crimea (Gromova, 1932a; Gromov, 1948; 
Pidoplichko, 1951). The southern distribution limit of P. spelaea Goldf. 
was probably somewhere in the Mediterranean area. Remains of Paleolithic 
cat from Lebanon, Syria and Palestine probably belong to the real lion 
(Picard, 1937). There have been no findings of P. spelaea Goldf. in 
Iranian caves. On the Caucasus remains of a large Pleistocene cat were 
discovered in the Sakazhia cave, in loamy talus deposits in the vicinity 
of Sochi and in asphalt sands of the Apsheron Peninsula (Figure 111; Map 33). 

Gromov's assumption (1948) that Panthera spelaea became 
extinct on the Caucasus just before the Bronze Age, which would have been 
later than in other regions, is open to question. In all probability, it 
became extinct simultaneously with the cave bear, woolly rhinoceros and 
mammoth. Related species, such as lion and tiger, appeared on the 
Caucasus as postglacial immigrants from the south. 

Lion — Panthera leo L. The subject of the lion's place in the • 
Caucasian Holocene fauna and of its relationship to the Pleistocene 
Panthera spelaea is very complicated. The majority of paleontologists 
considered P. spelaea the direct ancestor of the contemporary lion. 
Without going into details, Pidoplichko (1951) claimed that the large 
Pleistocene cat was also a lion. However, as our investigations have shown 
(Vereshchagin, 1951b), P an t h e r a spelaea possessed more specialized 
feline cranial features than the contemporary African lion. 

A number of West European paleontologists have recorded lion remains 
in caves in England, France, Spain, Germany, Italy, Poland, Sweden, 
Czechoslovakia, Hungary, Morocco and Rhodesia, mostly from Holocene 
strata (Wolf, 1938, 1939). 
271 Lion remains were reported by Picard (1937) in Acheulean and 

Mousterian strata in Palestine. 

Gromova (1932a) mentioned teeth and metapodia of lion found in an 
ancient Greek cave in Olvia; however, these remains may have come from 



266 



a lionskin rug brought from the south and not, as reported, from an animal 
caught on the north coast of the Black Sea. 

Spendid low- and high-relief representations of lions found during the 
excavations of ancient Sumer, Babylon and Nineveh verify that the lion was 
very common in southwest Asia in the third to first millennia B.C. Another 
series of representations of lions executed on stone by the Hittites in the 
central part of Asia Minor was published by Osten (1929-1930). 




FIGURE 111. Skull of P an thera spelaea Goldf. (female)from 
Binagady asphalt 

Lions are known on the Caucasus only from pictures and legends. Among 
them are the following noteworthy findings: relief representations of two 
lions on the gold plating of a beaker from the Trialet mounds, 2nd century 
B.C. (Kuftin, 1941); contour images of two lions on a gold cup from a 
Kirovakan burial, second millennium B. С . (Piotrovskii, 1949); contour 
images of lions on rocks in Kabristan from the first (?) millennium B.C. 
(Vereshchagla and Burchak-Abramovich, 1948); contour images of lions 
on silver containers and sculptures on gold plates (Figure 112) found 
near Maikop, second millennium B.C. (Farmakovskii, 1914); relief 
representation of a lion on the gold plating of a quiver from burials on the 
Taman Peninsula (Figure 113). 

Representations of lions and lionesses dismembering fallow deer and 
wild asses were found on a gold vial from the Solokh burial in southern 
Ukraine (6th century B.C. ). Mantsevich (1949) concluded that this vial was 
made by Ionian craftsmen, but (despite our advice) erroneously 
classified the wild asses as horses and the lionesses as leopards. 

We do not exclude the possibility that the lion penetrated into Ciscaucasia 
and the Russian Plain in the xerothermic period. Pogrebova (1950) formed 
the reasonable opinion that most of the artistic works and representations 
of animals found in the Scythian mounds were executed locally and were, 
272 therefore, modelled on the local fauna. The find, in the strata of ancient 
Tanais, of a plate of local raw limestone (43X56 cm) depicting a lion 
supports this theory. This stone is kept in The Hermitage (Knipovich, 1949). 

Later likenesses of lion appear on many relics of Armenian, Arabian 
and Iranian culture in Armenia, Azerbaidzhan and Georgia. Lion heads 
appear on the walls of the Old Derbent fortress which dates from 



267 



the time of the Arab caliphate, and on the fortress gate in Baku, which 
was erected in honor of Shah Abbas in the 17th century. 

Lion images are frequently found in the medieval art of the Dagestan 
mountain region (Bashkirov, 1931). In Iran at the time of the Safawids, 
hunting rugs were designed with lion images (Kverfel'dt, 1940). 





FIGURE 112. Gold plates with figures of lions from a Maikop burial 



Both the earlier and later representations of lions found on the Caucasus 
could be easily accounted for by the influence of Hittite, Assyrian, 
Babylonian, ancient Iranian and Arabian cultures, and even by the 
importation of finished articles from southwest Asia, But a comparison 
of the pictures on Kabristan rocks with Scythian articles of Ciscaucasia 
suggests a different explanation. 

It is possible that the distribution area of the European and of the Asian 
lion was irreversibly reduced in historical time even in southwest Asia. 

Herodotus and Aristotle mention the presence of lions in the Balkans 
in the second half of the first millennium B.C. The lion figures in Homer's 
Iliad. A Greek military division of the 5th century encountered north 
African savannah animals near the source of the Euphrates (Xenophon, 1896 
edit. ). These included ostriches and onagers which became extinct in the 
19th century. 

Surveys on the early distribution of the lion in the Near East have relied 
heavily on such literary references as Sundevall (1863), Keller (1909), 
Gromova (1928) and Korner (1930). 

Lucian of Samosata (125 A. D. ) reported a large cat (lion? I ) which jumped 
on the horse of a Scythian hunter (Latyshev, 1947-1948). Zhitkov refers 
to the probable presence of a large cat — "a ferocious aninnal" — on the 
southern steppes in old Slavonic tinnes in "The Lay of the Host of Igor" 
(Slovo о polku Igoreve, 1936, p. 229), as does Vladimir Monomakh in 
"Instructions" (Pouchenie). However, Sementovskii (] 857) understood 
"ferocious animal" to mean only a wolf. Moisei Kalankatuiskii wrote 
(lOth century A.D.; 1861 edit., p. 6) in what may be an imperfect 
translation: "Blessed is the country of Agvan='4 . . where are wild animals: 
273 lions, tigers, panthers, wild asses; and a great number of birds: eagles, 
falcons and their like." It is interesting that the reference to wild 
asses was verified by kulan bones found in 1946 during the excavation 
of a 13th-century settlement in Baku. At the time of the Arabian 
caliphate and later, eastern Transcaucasia was entirely suitable for 
the habitation of lions. The tugai on the Kura and Araks rivers and the 

* Ancient Albania and Azerbaidzhan . 



268 



vast reed-grown areas on the Mugan floodplains were densely populated by 
boar and deer and herds of kulans and goitered gazelles grazed in the open 
semidesert. 

Herds of ungulates were attracted in spring, summer and fall to the 
piedmonts grown with juniper — pistachio forests and a luxuriant grass cover. 

In the hunting ode "Shikhariya" dedicated to Manuchekhr, the son of 
Shirvan Shah, the poet Khagani (Avsaraddin), who lived in Shemakha 
(1120-1194), glorifies the Shah's lion hunt, mentioning the area between 
the Caspian Sea and Shemakha (History of Azerbaidzhan Literature, Vol. 1, 
1943; Text of Arasla in Azerbaidzhan). 




FIGURE 113. Part of gold lining of a quiver from the Semibratnoe site, 
showing a lion dismembering a deer 



In Farsi the word "shir" means lion and we can assume that the Shirvan 
region derived its name not only from the ruling shahs, but also from the 
lions which inhabited it. 

In his poems "Iskandarnama, " "The Seven Beauties" and others, 
Nizami Ganjawi (llth-12th centuries, A. D. ) constantly used the lion as a 
literary allusion. 

Usamah ibn Munkidh (1922-1923 edit. ) described lion hunts in Syria 
and Palestine in the 12th century when lions were as common there as hares 
in present-day Caucasia. 



269 



At the end of the 19th century, Blanford (1876) indicated that lions had 
become extinct in northern Iran, but were still living in Mesopotamia on 
274 the western Zagros spurs and southeast of Shiraz. The Persian lion, 

P. leo per sic a Meyer, 1826, was described from a Teheran specimen, 
i.e. , from northwestern Iran. 

Chardin records that lions and tigers still inhabited the forests of 
Mingrelia and Imeretia at the beginning of the 17th century (1735, p. 51). 
Chopin (1852), who lived in Armenia, stated that during his lifetime lions 
were no longer observed there, but tigers were occasionally seen wandering 
from the Araks River. Danford and Alston (1880) mentioned the 
extermination of horses by a preying lion in 1873 or 1874 near Bilecik on 
the Upper Euphrates. According to Kinnear (1920), lions were no longer 
seen in Asia Minor by the end of the last century, but they still inhabited 
Mesopotamia and Arabia. After large-scale deliveries of weapons to the 
Near East during World War I, the number of Asian lions decreased rapidly. 
In the opinion of Pocock (1930, 1939) and of Harper (1945), the rapid 
extermination of the Asian lion was entirely a result of European 
colonization and wanton hunting by English officers stationed in Asia, who 
killed many of these animals for diversion. 

Considering the above, we can assume that the lion inhabited Trans- 
caucasia, penetrating there from the south in the Holocene. After a gradual 
withdrawal to the southeast and southwest (Map 33), it disappeared from 
this region by the end of the Middle Ages. The reduction of population and 
range of the tiger and striped hyena presents similar histories. Both of these 
animals were observed in this region until the first half of the 20th century. 

Tiger — P a nt he r a tigris L. Apparently remains of true tigers are 
known only from the caves of China and Java (Wolf, 1938). There are 
indications that findings in France and Siberia probably belong to 
Panthera spelaea Goldf. Tiger remains have not been found in the 
central Asian Paleolithic caves of Teshik-Tash and Aman-Kutan. 

The absence of tiger remains in the Binagady deposit is important for 
an understanding of the history of the tiger on the Caucasus. 

Today the tiger is distributed mainly in southeastern Asia. Like the 
cheetah, hyena and lion, the present-day tiger is a postglacial immigrant. 
The tiger spread widely in a northwesterly direction from southeastern 
Asia, probably only in the Upper Holocene. It is significant that this animal 
is not represented in the art of ancient Mesopotamia, but was depicted by 
the Persians, although not as frequently as the lion. 

Tigers occasionally appear in ancient Persian miniatures. In the 1620's 
the Russian merchant Fedot Kotov colorfully described a tiger in the 
Shah's menagerie in Kazvin (1852). 

Reviews and critiques of the literature on the distribution of the tiger 
on the Caucasus at the end of the 19th and the beginning of the 20th centuries 
have been published by Dinnik (1914a), Satunin (1915a) and Ognev (1935). 

In 1947 the author presented a chart of tiger distribution on the Caucasus, 
showing locations where they had been captured over the previous ten years. 
Eastern Transcaucasia has been the extreme limit of tiger distribution 
up to the present time. 

The tiger was so common near Lenkoran in the I860's that some were 
killed there each year (Radde, 1899). The tiger was frequently encountered 
at the end of the 19th century on the Lenkoran Plain and in the mountain 
forests of Talysh where at least one tiger was killed every year (Satunin, 



270 



275 1915a). At the beginning of this century the tiger population began to 

decrease rapidly. In the 1920's (Figure 114) tigers were sometimes seen 
in eastern Transcaucasia, although they were not inhabitants of this region. 
Between 1920 and 1950 only two cases of tiger-pelt yield were recorded: 
one in 1922 near Tiflis and one in 1932 in Talysh (Map 33). 




FIGURE 114. Tiger in the jungles of Lenkoran in the first half of the 20th century 

Soviet officers who lived in Gilan and Mazanderan in 1942-1943 told us 
of repeated reports of tigers and of their own personal observations of tiger 
tracks and the dismembered bodies of boars and domestic animals on the 
forested northern slopes of the Elburz Range. 

Deforestation, the extermination of boars (the main food of the tiger) 
and tiger-hunting have created unfavorable conditions for tiger life on the 
Caucasus. 

Panther — Pant he г a pardus L. (s. lato). Remains of P. pardus 
were found in the Paleolithic strata of numerous caves in Africa and India. 

Remains of this species have been found in caves in Syria, Palestine and 
Lebanon in strata dating from Acheuleanto Neolithic (Picard, 1937; Bate, 
1937) and, in central Iran, it is recorded from Paleolithic strata of the 
Bisotun cave (Coon, 1952). Wolf in 1938 reported related fossil evidence. 

It is possible that a contemporary species of panther of African-European 
type inhabited Western Europe during the warmer phases of the Pleistocene. 
Panther remains have not been observed in the Pleistocene strata of 
Eastern Europe, which in our opinion indicates that the limits of the range 
were more southerly — at the Caucasus and Kopet Dag ranges — during the 
Pleistocene . 

A unique find of Pleistocene P. pardus remains was made in the 
Paleolithic strata of Kudaro I and II caves at the head of the Rion. 



271 




The remains of P. pardus found in Holocene localities may have been 
brought there from elsewhere, as the only other reliable reports are on 
276 findings from Greek and Roman strata of Olvia and of ancient Theodosia 
(collection of the Institute of Zoology of the Academy of Sciences in the 
Ukrainian S.S.R,) and on findings in the Caucasus from the Mesolithic 
of the Sosruko grotto (on the Baksan) and from the "Dlinnokryly" cave 
(in the Araks ravine). 

The absence of P. pardus from the Middle Pleistocene fauna of 
Apsheron indicates that it first appeared in central Caucasia, penetrating 

the area along with the mouflon and 
porcupine from the south. 

During early historic time, 
P. pardus was widespread. It was 
known by the Scythians on the 
Ciscaucasian Plain. Bronze heads 
of P. pardus have been found in the 
Scythian hills of the Russian Plain 
and in Ciscaucasia (Figure 115). 

A golden figurine of a reclining 
panther was recovered from the 
Kellermess hill. A splendid small 
bronze statue of a panther was recently 
found in Samtavro excavations, west 
of Tiflis (cf . reproduction in: Vestnik 
drevnei istorii. No. 3, 1948). 

At the beginning of the 20th century, 
the distribution of panther was limited 
by the foothills and mountain regions 
of the Greater and Lesser Caucasus, 
and by the lowland of eastern Trans- 
caucasia (Dinnik, 1914a; Satunin, 1915c). During the last 50 years the 
number of panthers has rapidly decreased. 

From 1894 to 1898 eleven panthers were captured in the Kuban hunting 
regions. In 1904 only two panthers were caught there, while in 1905 and 
19 06 none were found. The panther is rarely encountered on the Main Range 
in our century. In 1912 two panthers were killed in the Zakataly region; 
after that time, despite the abundance of goats, gazelles and deer, only 
occasional panther tracks were seen in the ravines of the left tributaries 
of the Alazan. In North Ossetia two panthers were killed at the beginning 
of the 'twenties . In 1 949 one panther was killed in the gully of the right tributary 
of the Sunzha and in 1952 another panther was killed near Staliniri in 
South Ossetia. 

Panthers have been more numerous and more enduring in Talysh and 
Karabakh. In the middle of the last century Radde ordered and received 
twelve fresh panther pelts from the Lenkoran District over a seven-week 
period. From 1930 to 1940, i.e., about 70 years later, the Azerbaidzhan 
pelt -supply base received only two panther pelts and a small number of pelts 
was acquired privately by individuals. 

Not more than ten panthers were killed on the entire Caucasus in the 
1930's, and those mainly on the Zangezur ridge and in Talysh. Regular 
appearances of panther in the reed- grown thickets of the central Mugan, 



FIGURE 115. Bronze heads of P. pardus from 
Ukraine burials 



272 



for example on Makhmud-Chala Lake, were recorded from 1929 to 1932. 
Near the village of Kharmandali, a dead female panther apparently buried 
by the male was found in 1929. A male panther, which had fallen into a pit 
during a snowstorm, was killed near the village of Mashtagi on the Apsheron 
Peninsula on February 15, 1946 (Burchak-Abramovich andDzhafarov, 1949). 
Heavy snowfalls and storms have driven the panther away from the 
mountains. 
277 After World War II, encounters with panthers became more frequent 

and pelt yield increased. Fresh panther pelts were received from the 
Talysh Mountains, the Nakhichevan area and the Karabakh Upland. Thus, 
in Azerbaidzhan in 1947, the yield was five pelts and in 1948, thirty- six pelts. 

This population increase is an apparent result of the migration of this 
animal from southwestern Iran. Present-day panther distribution on the 
Caucasus is limited by the forests of the Talysh Mountains and the ridges 
of Karabakh, Zangezur and the Nakhichevan area (Map 34). 




FIGURE 116. Three-month-old panther from Talysh, showing 
typical spottiness of the young animal 

Photograph by author, 1947 



The panthers which inhabited the Greater Caucasus belonged to the 
pale -colored subspecies Panthera pardus с i s с au с a s i с u s , now 
nearly extinct. 

The Transcaucasian P. pardus tullianus Val. (Figure 116) is 
brighter colored with an admixture of rust tints and shorter fur. The 
panther which was killed in Apsheron belonged to this subspecies. The 
rapid disappearance of panthers from the Greater Caucasus in this century 



273 



is undoubtedly the result of persistent hunting of these animals. Unless 
panther hunting is forbidden, the animal will disappear in the next few 
decades from the Caucasian Isthmus. 

Lynx — Fe lis lynx L. (s. lato). A species which is closely related to the 
subgenus F. lynx pardina, dating at least from the Middle Pliocene, 
has been found in Upper Tertiary deposits of southern Europe (in Moldavia 
and in catacombs in Odessa, France and Italy). A cat mandible, similar 
in dentition to that of a small lynx, was found in Lower Pliocene deposits 
near Stavropol. 

Lynx fossils of the present-day type are characteristic of Lower and 
Upper Paleolithic settlements in the forest and forest-steppe zones of 
Western and Eastern Europe (Wolf, 1938, 1939; Gromova, 1948; 
Pidoplichko, 1951). On the Caucasus they have been found in the Sakazhia 
cave in western Transcaucasia. 

There are no reports of lynx remains from the Paleolithic of the eastern 
Mediterranean (Lebanon and Syria: Picard, 1937; Bate, 1937; Iran: Coon, 
1952). Thus, it can be assumed that in the Pleistocene the lynx was not 
distributed very far south of southwest Asia. Today the lynx range lies in 
plains and mountain forests and is limited by the polar circle. 

During the Holocene the lynx population was considerably reduced by 
hunting, but even more by deforestation. The lynx does not inhabit the 
riverain forests of the steppe zone nor the region to the south, despite the 
abundance of food, e.g., hare, beaver and roe. Perhaps the flooding of 
the area in the spring accounts for their absence; it may also account for 
the rarity of lynx remains in deposits found along the lower reaches of the 
Ural, the Volga and the Don. 

The present range of the lynx F. lynx orientalis Sat. on the 
Caucasus includes the entire forested area of the Greater and Lesser 
Caucasus and of the Talysh area (Dinnik, 1914a; Satunin, 1915a; 
Vereshchagin, 1947d) (Map 35). In adjacent areas of Turkey and Iran 
the lynx is encountered in islets of mountain forests. Generally this cat 
is found in mountain regions even though small shrubs may be the only 
vegetation. For instance, on the eastern limit of the Main Range many lynx 
live in the shrub thickets and undergrowth near the villages of Kyzyl-Burun 
and Khizy where hares are numerous. The lynx is also seen sometimes in 
woodless Dagestan and in small islands of oak forest in the Armenian 
Highland, e.g., along the headwaters of the Nakhichevan -Chai near Shalbuz. 
The lynx originated in the ancient Mediterranean area and today still inhabits 
the nearly woodless landscape of Palestine (Bodenheimer, 1935). According 
to Pidoplichko (1951), the lynx does not always inhabit forest land in 
southern mountain countries. 

The recent yield of lynx pelts from the Caucasus, from Ciscaucasia and 
from Transcaucasia has fluctuated widely. Significant increases in 
population were noted in 1928-1929, 1932-1933, 1938 and 1950 (Graph 15). 
The difficulties of mountain trade curtailed lynx extermination. 

European wildcat — Felis silvestris L. Remains of small cats 
(F. pygmaea Lartet, F. media Lartet, F. attic a Wagner and other 
species) closely related to the European wildcat are known from Miocene 
and Pliocene deposits in Italy, Austria and Greece, and are particularly 
associated with the Hipparion fauna. Fossils of a contemporary type 
are common in Pleistocene and Holocene strata of many caves throughout 



274 



Western Europe, excluding the Scandinavian Peninsula (Trouessart, 1898- 
1899a; Wolf, 1938). They are frequently found in Middle and Upper 
Paleolithic strata in Crimean caves (Puzanov, 1929; Gromov, 1948). On 
the Russian Plain a mandible of a small cat was recovered from Pleistocene 
deposits of the lower Don (Tsimlyanskaya). On the Caucasus several bones 
of this cat were found in Paleolithic strata of the Akhshtyrskaya and 
Tsebeldinskoe caves. There are records of Felis silvestris L. 
remains in the southeastern Mediterranean area taken from Acheulean to 
Neolithic strata in caves in Palestine and Syria (Picard, 1937; Bate, 1937). 

We found Holocene remains of European wildcat only in the Pyatigor'e 
area. It is interesting that only remains of domesticated cats were found 
in the medieval strata of the Sarkel fortress on the Don. It would appear 
that the range of the Caucasian subspecies, F. silvestris caucasicus 
Sat. , was not associated during the Holocene with the European subspecies 
in the Russian Plain. 

Today the range of the European wildcat includes Western Europe, Asia 
Minor and the Caucasus. It probably became extinct in the Crimea during 
the Holocene. Within the Caucasus the distribution of the European wildcat 
gradually declines eastwards (Map 36). There are no documented data 
from the Talysh and Elburz mountains; there is, however, some ambiguous 
information concerning a kind of cat seen in the forests of Lenkoran, Gilan and 
Mazanderan (Blamberg, 1853; Smirnov, 1922; Bobrinskii, Kuznetsov and 
Kuzyakin, 1944) which has not been verified. Associations with forest 
biotopes are quite pronounced for this species, but in Ciscaucasia wildcat 
distribution is limited by the small number of surviving plots of shrubs, 
forests and reeds, such as those of the Kabarda lowlands, the valley of 
279 the middle Kuma, the Stavropol Plateau and the Kuban River delta. In the 
southern part of the country, wildcat is more likely to be found in stands 
of forests, while on the woodless upland of the Lesser Caucasus it is 
practically nonexistent. In western Transcaucasia wildcats are numerous 
in the beech forests of the Adzhar-Imeretia Range and in the alder swamps 
of Colchis. Contrary to the studies made on the plain by Formozov (1946) 
(and repeated by Pidoplichko in 1951), the distribution pattern of wildcat 
does not show a relationship to the varying depths of snow cover on the 
irregular mountain terrain. 

The largest population of European wildcat is observed in the lower third 
of the forest zone; the animal prefers the beech— hornbeam forests. 

Caucasian Reservation data show the following distribution of 43 
encounters with wildcat: 76.7% in broadleaf forests, 18.6% in dark 
coniferous forests and 4.7% in the subalpine zone (Teplov, 1938a). The 
principal pelt yield on the Caucasus is obtained from submontane forests. 

According to paleontological data, the European wildcat is an autochthon 
of the Mediterranean area and, on the basis of present distribution patterns, 
can be regarded as an early Pleistocene settler from Europe which 
penetrated the Caucasus via the Balkans and Asia Minor. 

The hunting of European wildcats for their pelts during the last 25 years 
has not affected this animal's population (Graph 16). 

Jungle cat — Felis chaus Guld. The remains of this cat are known 
from Pleistocene strata of India (Wolf, 1938), According to Picard 
(1937), fossils have been found in Syria and in Palestine in deposits from 
the Mousterian to the Neolithic. Gromov (1937, p. 86) is of the opinion that 



275 



Azilian and Tardenoisian strata of the Crimean mountain regions may well 
contain jungle cat fossils. The finds in Western Europe (e. g. , in Stuttgart 
travertines) are insignificant. Pleistocene and Holocene remains of this 
species are unknown on the Caucasus, and finds of Holocene remains 
on the Russian Plain are also doubtful. In the 8- 13th century strata of 
the Sarkel fortress near Tsimlyanskaya two inferior humeral epiphyses 
were found which were similar in size to those of the jungle wildcat. 

In the 14th-century strata of a Slavic village in Poltava an inferior 
humeral epiphysis of a cat resembling this species in size was reported 
(Gromov, 1948), but the authenticity of this report is questionable. Our 
later investigations of cat fossils indicate that Felis lybica lived in 
the Crimea and on the southern Russian Plain throughout the Quaternary, 
but Felis chaus did not. It is probable that F e 1 i s chaus only 
penetrated eastern Ciscaucasia and the southern Russian Plain late in the 
Quaternary (during the Holocene). 

At present the main distribution area of Felis chaus is in southern 
and southeastern Asia, while on the Caucasian Isthmus the range is 
"undeveloped. " This cat inhabits eastern Transcaucasia from the Caspian 
Sea to an altitude of 900-1,000 m. In the west it inhabits the valleys of Kura 
to Gori; it has not been observed beyond Surami. In the Araks River valley 
the cat is encountered as far as the longitude of Mount Alagez. It is common 
along the Caspian shore, especially near Samur and along the lower reaches 
of the Sulak and Terek rivers, penetrating farther through the reeds along 
the Berovskie knolls and to the Volga delta (Dobrokhotov, 1939). 

In eastern Ciscaucasia this cat is found as far as Grozny in the Sunzha 
River valley and as far as Mozdok in the Terek River valley (Map 37). 
280 Felis chaus is not encountered at present in western Ciscaucasia. 
The pelts obtained from the Krasnodar Territory were brought from the 
southeast. In the scientific collections from this region, the jungle cat 
is not represented; although the literature, beginning with Bogdanov (1873), 
mentions the presence of this cat on the Kuban delta, it is yet to be 
confirmed. 

The data of Formozov (1946) on the habitation of this cat along the reed- 
grown banks of the lower Kuban River are erroneous. The finding he cites 
of a pelt and skull in the Cossack village of Grivenskaya was really of 
European wildcat. 




FIGURE 117. Jaw of cheetah from Binagady asphalt 

276 



The greatest number of this species is observed in the lowland forests 
of eastern Transcaucasia. Its favorite biotopes are reed— cattail thickets 
beside lakes and marshes and tugai thickets along rivers and rivulets. 
Here the cat hunts water voles and marsh birds. It is not encountered 
in mountain forests, but will migrate through open places in the 
Transcaucasian semidesert. 

Some unknown barriers impede its penetration into western Trans- 
caucasia, At the first glance it would seem that, like the jackal, the jungle 
cat would prosper in the Colchis, which has a topography similar to that 
of Girkan, a warm climate and many swamps. The absence of the cat from 
the Colchis cannot simply be explained by the "newness" of the distribution 
area on the Caucasus. After all, the Surami Pass is sufficiently low to 
permit passage, and predatory communities are known to settle rapidly. 
The absence of jungle cat remains from the Binagady locality, the "under- 
development" of its range on the Isthmus and, finally, the stenotopic 
character of the animal and its relative adaptation to warmth suggest that 
it is a postglacial settler of southern origin. The area of land on the 
Azerbaidzhan Plains inhabited by the jungle cat increased with the 
development of a local irrigation system and decreased with its 
abandonment. Most recently, the destruction of tugai forests and of bamboo 
groves and accelerated trade have caused a further decrease in range and 
population. It is difficult to discover the population dynamics of the jungle 
cat from figures on pelt yields because the pelts of this species and those 
of European wildcat are reported together. 

The granting of bounties on cats sharply increased the pelt yield. This 
reflects a state of species well-being, although nutria breeding is thereby 
endangered. 

Cheetah — A с i n о ny X jubatus Schr. Cheetah remains are known 
from the Upper Pliocene of southern Europe and Africa (Simpson, 1945). 
281 The species is very rare in Quaternary deposits and until recent times 

was known only from the Upper Pleistocene of China (Pei, 1939). In the 
Caucasus (Figure 117)they were first revealed in Middle Pleistocene strata 
of the Apsheron Peninsula (Map 38). At that time the cheetah (Figure 118) 
could have hunted saigas, asses, horses and hares. Possibly its distribution 
in the Pleistocene extended to eastern Ciscaucasia. The contemporary range 
of the cheetah includes Africa and southern Asia as far as China. Its 
preferred biotopes, according to Roosevelt, Pocock (1939, 1941) and other 
naturalists and hunters, are low -lying plains and hilly savannahs, deserts 
and semideserts where the animal shelters among rocks and shrubs. In 
Turkmenistan and Iran it lives in the open sandy, clayey and stony desert, 
but prefers river valleys grown with tugai thickets where it hunts gazelles 
and hares. 

There are no documentary data on the occurrence of the cheetah in the 
Caucasus in historical times. It was depicted (Figure 119) on an ornament 
of a silver container found in the Maikop burial (Farmakovskii, 1914). The 
wearing of a collar suggests the possibility of cheetah domestication by 
the time of the Scythian. On a 12th-century ceramic from the excavation 
of ancient Gandzha [now Kirovabad] there are pictures of cheetah somewhat 
reminiscent of those depicted by the ancient Egyptians. 

The Georgian Chronicles "Kartlis Tskhovreba" place the cheetah in 
eastern Georgia (Kartlia) in the Middle Ages and, from this source, its 



277 



presence was incorporated into the local fiction. Some local authors made 
extraordinary statements regarding its habitation of high mountains, e.g. , 
K. Gamsakhurdia in "The Hand of a Great Master" (1945). Usamah ibn 
Munkidh (1922-1923 edit.) mentioned that Asia Minor gazelles were hunted 
282 with cheetahs in the 12th century in Syria and Palestine. In the Russian 

Chronicles and also in "The Lay of the Host of Igor" (Slovo о polku Igoreve) 
"parduses" are mentioned — hunting "leopards" (cheetahs) — which served 
the Russian princes in the hunt during the llth-12th centuries (Zhitkov, 
1936). This practice was borrowed from the Persians or the Polovtsian 
khans, according to Zhitkov. This is quite probable, because nomads of 
the Russian Plain maintained military, cultural and commercial relations 
with those of Ciscaucasia and Transcaucasia from the time of the Bronze Age, 





< 







FIGURE 1 18. Cheetah in the Pleistocene semidesert of eastern Transcaucasia 



After Aristov (1866), Pidoplichko (1951, p. 178) erroneously ascribed 
these references to "parduses" (cheetahs) in the old Slavic records to the 
leopard, an animal unsuited for this hunt. The Azerbaidzhan khans and 
Armenian and Kartlian princes hunted with trained cheetahs up to the 14th 
century. In 1474 Josaf Barbaro saw the hundred hunting cheetahs of an 
Armenian prince. 

Adam Olearius (1870 edit.) recorded that a Persian shah 
hunted onagers using cheetahs in the Isfahan hunting grounds. Olearius 
wrote (page 935) in connection with this: "There* are numerous predators, 
such as tigers, leopards, wolves and bears. Tigers=''-"'= are so numerous 
there that they are sold in herds of 10-20 animals. The tigers were 
generally used for the hunt because they were rapid runners and became 
so highly domesticated that the hunter could place the tiger on the horse 
behind him. " Later information appears on the cheetah in Mazanderan and 
throughout the Caspian forests (Filippi, 1865, Blanford, 1876). 

' In the Gilan area . 
" Cheetahs. 



278 




FIGURE 1 19. Representation of cheetahs on a silver container from Maikop burial (according to 
Farniakovskii, 1914) 

In Asia Minor and Arabia the cheetah was rare at the end of the last 
century (Danford and Alston, 1880; Aharoni, 1930; Bodenheimer, 1935). 

The cheetah probably remained in a wild state in the Kura- Araks lowland 
and in the niiddle Araks valley until the l8th century. This is convincingly 
shown by the sudden decline in cheetah range and population in present-day 
Turkmenistan and its recent presence in northern Iran (southern 
Azerbaidzhan). 

The disappearance of the cheetah from Transcaucasia and southern 
Asia is explained by the depletion of steppe ungulates and the intensive 
pursuit of young cheetahs for training. 
283 By tracing the contemporary range of the cheetah in Asia and Africa, 

Harper (1945) established the reduced area of distribution and the population 
decline of this peculiar cat. 



Order LAGOMORPHA 

Lagomorphs appeared relatively early in time. Some genera of this 
order, for instance Mytonolagus and Shamolagus, are known from 
the Eocene. In the Mediterranean, central Asia and Mongolia, Tertiary 
lagomorphs are represented by pikas and hares (Trouessart, 1904-1905; 
Simpson, 1945; Borisyak and Belyaeva, 1948). 

On the Caucasian Isthmus remains of only two genera, Ochotona 
and Lepus , are known from the Lower Pliocene. 



Family LEPORIDAE 

European hare — Le pu s europaeus (s. lato). The earliest 
fragmentary remains of small hares were found in Pliocene alluvium near 
Stavropol and, later, in Pliocene fluvial deposits of the middle Zanga, 
north of Yerevan. 



279 



Numerous L. europaeus remains from the Middle Pleistocene of 
eastern Transcaucasia (Binagady) belong to a species closely resembling 
the conteniporary animal. 

The Upper Pleistocene remains of European hare were found in deposits 
of the lower Don. In Holocene deposits on the Caucasus, European hare 
remains are often found in caves used by eagle owls, as well as in strata 
of human settlements. 

The contemporary European hare inhabits almost all of th e С auc as ian 
Isthm us, frorn _where"TF "spre ac TTo tHe~"rioTt'h"ang " "fq Uve^ssjuth . The southern , 
UmjJ;_j3i4ts-pe«ge-j£~ unde line d_ but probably, extends to Mesopotamia. 

The pelt yield from European hare on the Caucasus shows the following 
distribution per 1,000 km , according to state pelt-supply records over 
several years (Vereshchagin, 1947): northern Caucasus — 3,300; 
Azerbaidzhan — 770; Dagestan — 594; Georgia — 219; Armenia — 127. 

Because most of the pelts are sold on the spot, the number of European 
hare killed in Transcaucasia is actually much greater, especially in 
Azerbaidzhan. The yield on the northern Caucasus is mainly from Krasnodar 
Territory and the Dagestan plain. In Georgia almost all of the hares are 
obtained from the dry eastern portion (Map 39). 

The hare population of the alpine and subalpine meadows of the Greater 
Caucasus is insignificant. It is more common in the dry inner valleys 
of the northern slope of the range and on the steppelike alpine meadows of 
inner Dagestan. On the Gunib Plateau the European hare rests by day in 
relict pine and birch forests. 

This species is not found at all in continuous mountain forests of beech, 
hornbeam and fir. It is true that they penetrate the forest zone moving 
upwards from the subalpine meadows along woodless slopes and ridges 
and along denuded ravines, but this penetration takes place mainly as man 
has changed the landscape. 

The contemporary range of the European hare over broad areas of the 
Caucasian Isthmus shows upon examination that these animals generally 
284 inhabit dry uplands and moderately dry ravines with shrubs or sparse 

forests. In discussing the biotopes of the European hare, it is necessary 
to distinguish between its day resting places and its feeding grounds, which do 
not always coincide. The natural distribution of couching and fattening 
places has undergone marked modifications over the millennia wherever 
man has engaged in animal husbandry. 

On the Ciscaucasian Plain the European hare feeds and couches in the 
open virgin steppe only in spring and early summer. If the steppes dry out 
or are burned, the animals concentrate in the river valleys. In fall and 
winter they feed on windblown places, i.e. , on the hills and southern slopes, 
and couch in ravines and shrubs. 

In continuously cultivated regions and in regions of intensive grazing 
the distribution of the European hare is very complicated. Almost all types 
of continuously cultivated land in the Stavropol area divert the hares from 
virgin land, especially when the virgin land is used for cattle grazing. 

Alfalfa and grass seedlings comprise the basic diet of the hare; hares 
also feed on ripened field melons, wheat and barley grains in stubble 
fields — the latter a probable result of careless combine harvesting 
(Kolosov and Bakeev, 1947, p. 12). 

In winter the hare feeds on crops where it couches. It will often couch 
on fall-plowed fields, crossing the next forest belt if disturbed. Permanent 



280 



fattening and couching places in the forest belts are observed only where 
there are very wide stands of advanced age. 

The contemporary widespread distribution of the European hare in the 
foothills, especially on the Trans -Kuban sloping plain, is the result of man's 
activity, i.e., eradication of forests and swamp drainage. 

T he h ar^__ is rare in w estern Transcaucasia, inhabiting the driest plot s 
of Шаек Sea terraces and only the cleared^-forest.^fpgthill glades.in^humid-.. 
Colchis. 

'""Trreastern Transcaucasia the maximum European hare population is 
observed in the dry foothills and on the slopes of the Kartalinia, lora and 
Azdhinour plateaus living in vegetation of Christ's -thorn and beard grass. 
The hares are temporarily driven out of this type of land only in winter 
when large sheep herds and their dogs appear (Vereshchagin, 1942b). Many 
European hares live in sparse tugai thickets, particularly in the tamarisk 
zone. In the Alazan-Agrichai and Khachmas lowlands they inhabit places 
where the original forests have been cut and where they can couch along 
the edges of forest glades. In the flat semidesert of the Kura- Araks lowland 
the European hare is very rare even in saltwort — caper communities. The 
plowing of semideserts and the establishment of fields, gardens and oases 
immediately promotes the appearance of the hare. It is particularly 
common in vineyards and fallow fields. Alfalfa and cotton plantings also 
attract the European hare (except when the cotton fields are being watered 
and weeded). 

On the Apsheron Peninsula the hare is found in vineyards and sand dunes 
where it couches among the ephedra shrubs and sand fescue. In winter when 
the hare is hunted, and when sheep are grazing in the vineyards, it 
couches on plowed fallow land and in winter barley fields, which, because of 
285 their microrelief, afford better protection in the furrows or between lumps of 
earth than winter-sown land. In vineyards the hares eat the ripe grapes 
and gnaw on field watermelons when they are thirsty. 

Hunting during the last decades has influenced the development of the 
hare population on the Apsheron Peninsula. Individuals couch during the 
day on stone fences or on ruins of houses from which they have a clear 
view of approaching danger. 

On piedmont plains hares couch in dry riverbeds where they hide behind 
boulders in order to escape easily from jackals and foxes. 

In the Talysh upland and in the Araks valley the main natural biotopes 
of the European hare are bottoms and slopes of ravines grown with tamarisk, 
common buckthorn and buckthorn thickets. In general, the habits of this 
small southern European hare are similar to the central Asian Tolai hare. 

It is characteristic of the European hare, a species of southern, 
Mediterranean origin, that it reproduces the year round and maintains its 
largest population in the south, its litters decreasing proportionately to the 
distance it migrates northward (Vereshchagin, 1938b, 1942b; Kolosov, 
1941; Kolosov and Bakseev, 1947). 

The hare population is greatly affected by tularemia, epizootics, 
coccidiosis and, more rarely, early frosts (Kolosov and Bakeev, 1947). 
These population fluctuations affect the trade to some extent (Graphs 17-18). 

The prime cause of the decline in hare population in recent years is year- 
round poaching at night with automobiles. 

The cultivated terrain of the Caucasian Isthmus will promote an even 
more valuable trade in European hare in the future. 



281 



Order RODENTIA 

More than 40 species of rodents are actually known from Cenozoic 
deposits of the Caucasian Isthmus and the southern Russian Plain. Holocene 
fauna of the same areas account for no less than 50 species identified as 
belonging to the families Sciuridae, Castoridae, Myoxidae, Dipodidae, 
Spalacidae, Muridae, Cricetidae and Hystricidae. 

The features of the ranges and the representatives are examined below. 



Family SCIURIDAE 

Only four species of Sciuridae are known from the Quaternary fauna of 
the Caucasus. Of these, the marmot is extinct; two susliks and the squirrel 
still survive. 

Marmot — Mar mot asp. Accordingto the finds made in the Kudarolcave 
in 1957 a kind of marmot inhabited the Greater Caucasus in the Pleistocene and 
Holocene. Marmot bones were found in Upper Pleistocene deposits in the 
basin of the Khram River in the Lesser Caucasus near the village of 
Orozman. Some relatively "fresh" marmot skulls resembling those of the 
steppe marmot were obtained by geologists from the karst funnel in the 
Belaya river ravine (Map 40). It is possible that the Caucasian marmot 
became extinct in the northwestern Caucasus at the beginning of this century 
(Vereshchagin, Geptner, Stroganova, 1959). The find of a small suslik, 
Marmota (?) sp. , in lower Quaternary strata near Akhalkalaki is 
noteworthy. 

Little suslik — С it e llu s pygmaeus Pall. (s. lato). The origin and 
distribution of the suslik in the Caucasus have been studied by faunists, 
286 zoogeographers and parasitologists but remain unclear. More paleonto- 
logical, morphological, zoogeographical and ecological data are required 
for the task. 

The genus Citellus is known from the Miocene and Pliocene of North 
America and from the Pliocene of Eurasia (Simpson, 1945). 

Remains of marmots of C. pygmaeus type were determined by 
Buchner and also by Pidoplichko (1951) from Quaternary deposits at Nogaisk 
in the Zaporozhe Region. 

Upper Pleistocene remains of Citellus pygmaeus have been found 
only within the limits of the present-day range on the Eastern European 
plain. Vinogradov and Gromov (1952) distinguished an earlier form — 
C. pygmaeus musicoides — and a more recent form — C. pygma- 
eus caspicus. The identification points up the probable retention of some 
archaic features of the mountain subspecies. 

From Lower Holocene, Mesolithic strata, suslik remains are known from 
the first Baksan gorge. More recent bones of little susliks from eagle 
owl pellets were collected by the author on the outskirts of the Caucasian 
Range: in the upper reaches of the Kalaus, in the Stavropol area and in the 
vicinity of Makhachkala. 

The contemporary range of the little suslik includes Eastern Europe, 
the northern Crimea, the Ciscaucasian plains and the plains of northern 
central Asia within the steppe, semidesert and desert zones. 



282 



There are reasons to consider the Neogene of these areas as the origin 
of this species without resorting to Pidoplichko's hypothesis (1951) of an 
earlier migration from the southern Caucasus. 

In the 1950's, the southwestern limit of the little suslik on the Cis- 
caucasian Plain crossed the Manych depression in the lower reaches of the 
Yegorlyk and, following the Stavropol upland from the north, penetrated 
the upper reaches of the Kalaus far to the southeast near the village of 
Sultanskoe, 15-18 km from the Kursavka station. Farther eastward the 
southern colonies extended to the Kuma near Vorontsovo-Aleksandrovskoe 
and continued parallel to the latitudinal course of the Terek. From the 
Terek estuary to the north, the range of this rodent continued unbroken. 
An isolated colony is located between the lower reaches of the Terek and 
the Sulak. 

On the right bank of the Sulak the suslik is dispersed over the foothills 
from Chir-Yurt to Buinaksk, while on the maritime terraces on the Caspian 
it penetrated southward nearly to Kayakent. 

Suslik populations exist in more isolation from the main basin of the 
range in the latitudinal valley between the Sunzha and Terek ranges, west 
of Grozny, and deeper in the ravines of the upper reaches of the Kuban, 
Malka, Baksan, Chegem and Bezingi Cherek rivers (Map 41). 

The territory between the Sunzha and Terek rivers inhabited by the suslik 
in the late 1930's comprised four sections with a total area of approximately 
1,000 ha (Ognev, 1947). 

The little mountain-dwelling suslik was recorded by Menetries (1832) 
who noted the animal on the northern slope of El'brus; the range was traced 
and described by at least 14 zoologists from the end of the last century to 
the first half of the present one. 

According to Sviridenko (1927, 1937) the westernmost colonies of the 
mountain suslik were disposed along the Uchkulan ravine and its tributaries. 
The uppermost somehow did not reach the estuary of Makhar and Dzhalp-Kol, 
and down the ravines they spread no farther than the Khudes estuary. 
287 The largest habitat was and is in the upper reaches of the Malka and its 
tributaries: Khasaut, Garbazi and Bolshoi Taluko. This is a vast plateau, 
gently sloping north and east from altitudes of 1,500-2,500 m. Sviridenko (1937) 
followed the junction of this range with that of the right bank of the Malka 
at an altitude of 3,072 m through the Buruntash Pass where the animals live 
not far from the El'brus glaciers. From here the susliks probably entered 
the Uchkulan basin. 

In the 1930's the suslik spread along the Khasaut to Narzan, a natural 
boundary, and to the Taluko Basin and the southwestern slope of Mount 
Kinzhal but not down into the Malka valley. 

In the east the suslik colonies are at present dispersed along the ravines 
of the Kurtyk and the Dzhovurgen, which are the left tributaries of the 
Baksan. In the Baksan ravine, the suslik is found 1.5-2 km above the gorge 
cut by the river in the Skalistyi ridge. From here, the suslik steadily 
spread in 1952 along both slopes of the ravine as far as the village of Nizhnii 
Baksan (Figure 120), and is especially numerous near the village of Byllym. 

Above Nizhnii Baksan there is another valley colonized by susliks which 
is separated by 8 km from the nearest colony on the left slope of the ravine. 
The animal's penetration into the Verkhnii Baksan ravine was probably 



283 



accomplished by way of the Kurtyk River ravine, which is the left tributary 
of the Baksan from the west. In the upper trough of the Baksan the species 
was observed in 1952 in the Tegenekli area and even somewhat higher, 
although Sviridenko (1937) did not find suslik even at the Adyly-Su estuary. 




FIGURE 120. Habitat of the little suslik in the Baksan ravine 

Photograph by author, 1952 

Through the valley of the Dzhigiom River, the right tributary of the 
Baksan, the suslik penetrated the Chegem valley in which it lives at a very 
288 high altitude especially on the Bashi-Buzu-Su tributary. From the northern 
slopes of Mount Chegem-Bashi it spread into the Bezingi Cherek 
ravine. 

In the Balkar Cherek, the suslik has not been encountered, although there 
are places suitable for it in this area, for instance, near the villages of 
Kunim and Zilga. 

The suslik has never been found in the Urukh ravine. In 1947-1948, the 
author travelled through the entire Urukh area, including its tributaries, 
through the Balkar Cherek and the Cherek Rion pass without finding any 
suslik, and local inhabitants reported that suslik had never been seen in 
their area. Dinnik did not find them even in 1883 ( 1884a, 1884c). Thus the 
Rossikov report (1887, p. 44) of encountering a mass of susliks in 
mountainous Digora is probably purely imaginary. Ognev, citing Rossikov 
(1947, p. 139), added confusion by placing Mount Kun'-Ityg and the 
Zhelanoko ridge (i.e. , Dzhinal) in North Ossetia, whereas they are located 
in the Malka Basin (i.e. , the Kabarda piedmont). 

The present geographic variations of the little suslik range within the 
Caucasian Isthmus (Chapter 4) does not account for the isolation of separate 



1704 



284 



parts of the early range of this animal on the Caucasus, because we simply 
do not know how rapidly the formation of separate features of the species 
in different environmental conditions takes place. 

A study of suslik distribution throughout its total range clearly shows 
the early adaptations of the species to artemisia semidesert. It is precisely 
in this zone of artemisia semidesert, clayey soils and specific plant 
groupings that the greatest suslik population density is observed. 

Considering the Neogene origin of the artemisia semidesert landscape 
on the northern margins of central Asia (L.avrenko, 1938) and the mountain 
xerophytes of the eastern Mediterranean (Bush, 1935; Grossgeim, 1948), 
we may assume that the age of the indicated adaptations is at least Upper 
Pliocene . 

Suslik penetration into the true artemisia steppe is only spotty. Birulya 
(1941) assumes that the factors limiting its distribution are the high, closed 
stands of homogeneous grass, the heavy spring thaws and the stagnant, 
overheated air in the feathergrasses. 

He further stated — and correctly so — that during the course of evolution 
the animal adapted to open biotopes. On the level steppe with a low grass 
stand, the colony could maintain prompt signalling of mutual danger, a 
complex system of burrows and normal heat exchange conditioned by moving 
ground air. It is significant that among mountain susliks of the Caucasus 
these early adaptations have not disappeared but have only changed somewhat. 

Susliks settling in the mountains on the moraine or on alluvial gravel 
have adapted to burrows dug under large boulders, thereby preventing the 
collapse of the burrows in the sandy ground and utilizing the friable spaces 
between cobblestones. 

On mountain-steppe meadows the suslik generally lives in colonies in 
open places near rock outcrops and digs burrows through the vertical walls 
of paths worn by cattle grazing on the slopes. This habit of digging burrows 
in different kinds of vertical walls is also characteristic of the Volga-Don 
steppe suslik. 

The arid floors of the Baksan and Chegem ravines have environmental 
conditions resembling those of the desert -plain and for this reason are 
289 more constantly and densely populated by susliks. The immigrating 

population decreases rapidly in proportion to the altitude. Thus, on Baksan 
near Byllym (August 1952) we obtained the following data on suslik population 
by altitudinal zones (Table 69). 

TABLE 69. Density of the suslik population in the Baksan ravine 



Artemisia- thymus 

grouping on the lower first 

terrace and in places 

where there are small 

stone heaps, stone fences 

and weeds; altitude 

1.100m 



Festuca-artemisia 

grouping on the upper, 

third terrace on stone 

placers; altitude 

1,200 m 



Subalpine meadow with 
motley-grass, in some 
places with outcrops; 
southern exposure of 

slope approximately 30°; 
altitude 1,850 m 



Number of burrows per 
ha 



76 



25 



285 



Sviridenko's data (1937) on the burrowing habits of the suslik can be explained 
only by his tendency to ennphasize the morpholgical-ecological isolation of 
the mountain suslik from the plain suslik. Thus, he claims that susliks 
will tolerate arboreal habitats, burrowing deeply in forest glades, but 
digging shallower burrows in sun-warmed artemisia-grown ground. 

The ages and causes of isolation of suslik colonies in Ciscaucasia are 
various. The isolation of the extreme southeast colonies in the Dagestan 
piedmont is caused by the moi-e or less constant flow of the lower courses 
of the Terek and Sulak rivers. Their frequent bifurcation and meandering 
create fairly wide land areas, grown over with tugai forests, and reed-grown 
lakes which are ecological barriers for susliks. When a river changes 
course, the habitat of the animal population previously located on one bank 
is left on the other bank. It has also been experimentally proved that susliks 
can swim across narrow rivers (to 150 m wide) and can cross them on 
bridges (Sviridenko, 1927, 1937). Isolated instances are known of gravid 
susliks being transported alive over considerable distances by birds of prey, 
such as Larus argentatus. 

For various reasons, the isolation of suslik colonies located between the 
Sunzha and Terek ranges seems to us to be quite ancient. In the first place, 
the susliks from the Alkhan-Churt valley are morphologically different 
from those of the Kuma and of Dagestan — they have been reported as 
subspecies C. pygmaeus boehmii Krass. Secondly, it is known that 
the Terek cut through the Terek gorge between Zmiisk and Terek at a period 
no later than the Middle Pleistocene. From the Malka estuary to the Sunzha 
es<-uary, the Terek is a powerful river with a strong and straight current 
and nearly devoid of oxbow lakes. Well-developed forest belts of old 
galleries unsuitable for suslik settlement are disposed along the Terek, and 
the northern slope of the Terek Range is covered by forests impassable to 
susliks. 

The eastern approach of susliks to the Alkhan-Churt valley was barred 
by a wide strip of broadleaf and tugai forests on the Gudermes sloping plain 
and by the Argun, Gudermes and Aksai rivers. 

Therefore the formation of the Terek-Sunzha colonies should be 
correlated with the formation of the festuca— artemisia and andropogon— 
290 artemisia steppe landscape of the Alkhan-Churt valley, i.e, at least with 
the Middle Pleistocene. 

The origin of the Elburz mountain colony of susliks is more enigmatic. 

Up to the present, Shchukin's hypothesis (1925) has served to a greater 
or lesser extent as the basis for various modifying theories on the 
penetration of suslik into the mountains. Shchukin differed his hypothesis in 
answer to statements of Kuznetzov (1890) and Krasnov (1893-1894) on the 
penetration of dry mountain vegetation along the interior, longitudinal 
valleys of the northern Caucasus to Dagestan. He implied the penetration 
of suslik from the plain along the transverse ravines after the withdrawal 
of the glacier on the periglacial steppe, and a subsequent cutting-off of the 
colony from the north by the establishment of a forest strip. However, the 
question of the origin of the mountain suslik cannot be resolved so simply, 
because in Elburz the animal inhabited not only the valley floors, but also 
the passes of meridional ranges. Besides which, the colonies disposed 
along the low ridges of the Malka Basin were not cut off from the north by 
a forest strip; the foothills of this region have always been woodless. 



286 



Still Sviridenko (1927) assumed that, as an "emigrant from central Asia," 
the little suslik could have appeared on the steppes of the Volga-Don and 
in Ciscaucasia only after Khvalynsk time. Even from our viewpoint, the 
fairly recent establishment of the suslik range in Dagestan is confirmed 
by the absence of the species from central Dagestan where living conditions 
are very favorable for it. Later, after studying the peculiarities of suslik 
distribution in the mountains, Sviridenko (1937) concluded that they appeared 
in the mountains of central Caucasia during the time of the third Pleistocene 
maximum giaciation (Riss) — i. е., long before the Khvalynsk transgression — 
thus "broadening the open-steppe ranges." Sviridenko's conclusions 
were based on Tutkovskii's rejected hypothesis (1909) of an arid climate 
in preglacial time. Contrary to this view, there is no place on the Caucasus 
where glaciers and foehns have not formed periglacial steppe, and 
contemporary glaciers are often embedded directly into pine and mixed 
forests (e.g. , in Karaugom and Tseya). The formation of a dry-steppe 
zone within a mountain system is associated with fringing, isolated ridges 
which act as rain screens, rather than with any glacial influence. For 
example, central Dagestan is very arid, but has no glaciers. What is of 
importance is Sviridenko's assumption of the survival of mountain susliks 
from the time of the major giaciation of the Caucasus, which contradicts 
earlier geologic and geomorphologic data to the effect that enormous glaciers 
covered places inhabited by susliks. Also of significance is Sviridenko's 
zoogeographical structure which coincides well with Bush's concept (1935) 
that the mountain-steppe vegetation of the longitudinal valleys of the northern 
Caucasus derives from the ancient mountain xerophilous vegetation of the 
entire Mediterranean area. 

loff (1936), who accepted Shchukin's hypothesis of the postglacial 
penetration of the suslik into the mountains, was inclined to explain the 
entire history of discontinuities in the Ciscaucasian range of the suslik by 
extreme fluctuations in population, colonization and extermination over 
great expanses. 

In point of fact, in the early 'fifties, the break in range between 
the southern suslik colonies in the upper reaches of the Kalaus 
291 and the Bermamyt slopes was only 92-95 km on a straight line and that 

between the Kuma Steppe and the Baksan ravine colonies was only 120 km. 
This break may mend because the distribution of susliks to the south has 
accelerated in the last decades under the influences of human activity and 
the naturally arid climatic cycle. Previously it was thought that the mass 
settling of susliks on the Caucasian steppes began, as it did in the former 
Astrakhan Province, about 1900-1901 (Pirkovskii, 1913; Sviridenko, 1927). 
However the suslik probably inhabited this area earlier, but was not 
reported because the country was so little cultivated and, mainly, because 
its presence was not investigated. There could also have been local 
population variations as reported by Lebedev (1912) and Satunin (1920, p. 81). 
All of this, however, neither excludes nor contradicts the established fact 
of the suslik's rapid advance southward in present time (Sviridenko, 1927; 
Romanova, 1936; Babenyshev, Birulya, Besedin, Golosovskaya, Egorov, 
Korf, Yanushko, 1937). 

In 1952, the author found susliks near Sultanskoe and Mironov. In the 
same year, Pavlov, Pushnitsa and Shiranovich together published a chart 
which indicates a shift of suslik range southward along the entire border 
for some kilometers. 



287 



The settlement is advancing southward at a rate of 2.5-3 km per year. 
At this rate, the plain susliks may reach the mountain colonies of the upper 
reaches of the Baksan and the Malka within 30-40 years! 

From geological data (Vardanyants, 1939, 1948; Nikolaev, 1941) we 
can assume that the suslik appeared on the Greater Caucasus at the 
beginning of the Quaternary when the recession of the Apsheron sea and 
a considerable peneplanation of the ranges created areas of the Caucasian 
Isthmus suitable for wide distribution of steppe and desert species of flora 
and fauna. By concentrating in those parts of longitudinal valleys and 
southern cuesta scarps with mountain-xerophilous vegetation, the suslik 
could have survived the maximum glaciation in the mountains, although 
the possibility of Pleistocene and Holocene migrations cannot be excluded. 

The later settlement of the suslik was dependent upon the animal's 
surmounting the ecological barriers which the forest and meadow zones in 
the Chernye Gory mountains and on the Pastbishchnoi and Skalistyi ridges 
presented. Suslik penetration into the mountains did not, in our view, follow 
the transverse ravines of the central Caucasian rivers, as proposed by 
Shchukin (1925). The ravines of the Skalistyi ridge east of Chegem have 
always been narrow and humid, and impassable to the suslik for distances 
of 8-10 km. These gorges became deeper with the tectonic cycles of the 
Pleistocene and attained maximum development in the postglaciation. 

Perhaps it is because of this barrier that the suslik is not found in the 
xerophytic zone of the upper reaches of the Balkar Cherek, Urukh, Ardon, 
Fiagdon, Gizel'don, Sunzha and Argun, nor in the whole of central Dagestan, 
although the living conditions are certainly no worse than those below 
El'brus. 

The suslik's presence in the Alkhan-Churt valley as contrasted with its 
absence in the valleys of the right tributaries of the Sunzha is especially 
significant. 

The suslik's relatively recent appearance in the mountains is indirectly 
indicated by the feeble development of the eastern range along the 
292 longitudinal valleys. The Balkar Cherek ravine is an example of an area 
which the suslik has not had sufficient time to populate. 

During the time of suslik migration from the plain to El'brus, the path 
of distribution probably lay east and west of the forested Pyatigor'e area, 
which was a barrier of sorts to animal penetration southward. It is likely 
that suslik bones deposited by eagle could be found in caves in this area. 

In 1952 we made a paleontological field trip into the Pyatigor'e area 
and found that no suslik bones had accumulated over the past 1,500-2,000 
years in the regions of Zheleznovodsk, Pyatigorsk and Kislovodsk. This, 
however, does not mean that there were no susliks in the Pyatigor'e area 
in the postglacial xerothermic epoch or in the Pleistocene, particularly 
since suslik remains have been discovered in Lower Holocene strata in the 
Baksan gorge. 

The question of the origin of susliks in the Caucasian mountains could 
possibly be clarified further by more intensive investigations of shelters 
and caves in the mountains of Razvalka and Verblyud, and the gorges of Kich- 
Malka. Kuma and Podkumok; the acclimatization of mountain susliks on 
the plain and of plain susliks in the mountains should also be investigated. 
During this study it should be possible to trace the rate of ecological and 
morphological evolution for both forms. For the present, the best estimate. 



288 



based upon acknowledged facts, for the settlement in the mountains of the 
suslik from the Ciscaucasian plains is the relatively Upper Holocene. 

In the Recent epoch, man, by his activity, has unknowingly contributed 
to the rapid settlement of the suslik in the mountains, as well as on the 
plains (see Chapter VI). Efforts at extermination and the pelt trade (Graph 20) 
were not enough to impede this process in its early stages. 

Modern extermination methods and accelerated cultivation of the steppes, 
the development of shelterbelts and of artificial irrigation systems will 
virtually eliminate the suslik from the Ciscaucasian plains. 

Asia Minor suslik — С i t e 1 1 u s citellus L. Fossils of suslik 
C. citelloides Kormos have been reported from Pleistocene strata in 
European caves (Wolf, 1939), and Citellus sp. from the Upper Paleolithic 
of Syria (Wolf, 1939). In the Caucasus, only more recent remains of 
C. citellus L. are known from Holocene alluvial loams in the Zanga 
ravine northwest of Yerevan (Dal', 1949a). 

The contemporary distribution of the southeastern subspecies 
C. citellus xant hopr у mnus Benn. includes the Armenian Highland 
and Asia Minor. 

In Armenia the suslik occupies a large area of the Araks valley, from 
the region of the Kara-Burun station and Alagez to the village of Illi, 
northwest of Leninakan; the upper valley and the Pambak River Basin to 
the Amamla station; and the western and northern slopes of Alagez (Map 41). 

The altitudinal distribution of the animal occupies a zone approximately 
1.5 km wide within an altitudinal range of 1,100 to 2,700 m (Dal', 1948b; 
Avetisyan, 1950). In the mountains, it lives in meadow -steppe, or even 
in subalpine meadows. In the Araks valley it is found in the mountain- 
artemisia semidesert, but not in the desert valley of the Araks. 

In view of the considerable ecological variability of the Asia Minor suslik 
(Avetisyan, 1950) it should be considered endemic to the submontane and 
293 mountain landscape of arid and relatively hot regions. The range area 
includes Asia Minor, the Aegean and the Balkans. The former habitat of 
the suslik in the Zanga ravine indicates a wider distribution during isolated 
epochs of the Quaternary. The present-day distribution in some areas 
suitable for the habitation of Asia Minor suslik was achieved only 
circuitously; for instance the arid Tertiary foothills bounded on the north 
by the Kura lowland were reached by way of land lying east of the Karabakh 
Mountains. The main causes of the feeble development of the Caucasian 
range of this species, in comparison with that of the steppe vole and the 
Asia Minor hamster, are its more conservative living habits and its lesser 
migratory ability. 

The local breaks and contractions in the range of this suslik on the 
Armenian Highland may be a result of orogenic processes and lava flows 
which covered large areas. At the time of cooling and glaciation, the upper 
border dropped at least 400-500 m. It is possible that the suslik also 
inhabited southern Dzhavakhetia in the Pleistocene. 

In the recent epoch the distribution of the Asia Minor suslik expanded 
in the upper zone and shrank in the lower as a result of pelt trading and 
chemical extermination (Avetisyan, 1949). 



289 



Family CASTORIDAE 

Castoridae remains are known from the Lower Oligocene of Eurasia and 
North America. In their adaptations the beavers greatly resemble the 
terrestrial porcupines and, to a lesser extent, the amphibious voles and 
Octodontidae. Miocene-Pliocene fossils of the genera Steneofiber, 
Amblycastor, T r о g on t he r iu m and Pliocene-Pleistocene fossils 
of the genus Castor are numerous along a wide belt of the northern 
Mediterranean and central Asia stretching from the British Isles to Mongolia. 

Fossils of four beaver genera, Steneofiber, Amblycastor, 
Trogontherium and Castor (Map 42) have been found on the Caucasian 
Isthmus. 

European beaver — С a s t о r fiber L. Remains of European beaver 
were first found on the Caucasus in Upper Pliocene conglomerates of the 
Tamian Peninsula. The Taman beaver, C. tamanensis N. Ver., from 
the Sinyaya gulley site, a natural boundary, was probably the direct ancestor 
of the contemporary European beaver (Vereshchagin, 1957a). Other 
Pliocene representatives of the family Castoridae — Steneofiber, 
Amblycastor and Trogontherium — also known from the Caucasian 
Isthmus, do not have direct affinities with European beavers. 

A European beaver of a completely contemporary type lived in the 
Caucasus during the Upper Pleistocene and the Holocene. Its remains are 
found in the caves of Sakazhia (Gromov, 1948) and Uvarova in the vicinity 
of Kutaisi, in the Neolithic strata of the Sagvardzhile cave, and in Upper 
Bronze Age strata in the Samtavro burial fields in the vicinity of Mtskheti. * 

Thus the beaver habitat in the Holocene on the Rion, Kvirila, Kura and 
Aragva tributaries has been documented (Figure 121). 



294 






FIGURE 121. Remains of European beaver 

1 — Femur from Samtavro deposits (first millennium B. C. ); 2 — pelvic 
fragment from the Neolithic strata of Sagvardzhile cave; 3 —pelvic fragment 
from the Paleolithic strata of the Uvarova cave 

Information on beaver fossils from Sagvardzhile and Samtavro was kindly provided by N.O. Burchak- 
Abramovlch, Doctor of Biological Sciences, 



290 



The remains of European beaver are very common in Quaternary deposits 
on the Russian Plain, but in the Crimea they have been found only in strata 
of the Upper Scythian epoch. South of the Caucasus, there have been no 
validated findings of beaver remains; Bate (1937) reports none from caves 
in Palestine, Syria and Lebanon. 

Beaver fossil identifications from Iranian caves (Coon, 1951) are doubtful 
because these fossils were probably mixed with porcupine remains. 

The great number of recent beaver remains in the lower Don floodplain 
(Sarkel near Tsimlyanskaya, Novocherkassk) indicates that the contemporary 
habitat of this animal on the rivers of the Cis- and Trans-Kuban plains is 
quite probable. Until the 18th century, perfectly suitable beaver habitats 
were located on the Psekups and on several small left tributaries of the 
Kuban and even of the Terek. 

Published data on the presence of beaver in the Caucasus during the 
present epoch are numerous and well-known. 

The beavers of Scythia and Colchis were often mentioned by ancient 
authors (Herodotus, Strabo). More recent references to beavers 
in Mingrelia and generally throughout Georgia can be found in Lamberti 
(1654) and in Chardin (1 686), in the Georgian Chronicles "Kartlis 
Tskhovreba, " and in the "Geography of Georgia" by Prince Vakhushti 
(1904), written at the beginning of the 18th century. At the end of 
the 18th and the beginning of the 19th century beaver pelts and 
"castoreum" were probably important to the import-export economy of 
Georgia. This is apparent from the status of Tiflis as a trade center in 1803. 

Giildenstaedt (1879) and Pallas (1831) recorded the probability of beaver 
habitation on the Kuban and the Sunzha. Menetries sent a beaver skull, found 
on the bank of the Sunzha, to Academician Brandt for his collection. Rovinskii 
(1809) mentioned that beavers were occasionally observed on the Kuban. 
Nordmann (l840) recorded several beavers killed in the 1730's in the Notanebi 
Basin and in Colchis, and indicated the beaver's presence on the Terek. 
Hohenacker (1837) mentioned the Araks valley as a habitat of beaver. 
2ПС Chopin'sdata (1852, p. 807) on beavers and "castoreum" obtained in 
Armenia, Turkey and Iran are questionable. 

All these data are to be found in Brandt's monograph (1855). 

The beaver skull preserved in the Zoological Institute of the Academy of 
Sciences U.S. S.R. under No. 6330 ("Menetries, 1831"), represented by 
Brandt (1855), has no trace of water sluice, river sand or silt, and has been 
thoroughly prepared and scraped with a knife. Perhaps this animal was 
killed not long before the arrival of Menetries in Sunzha, qr was gathered 
by him from a northern location en route to or from the Caucasus. 
Menetries himself did not mention (1832) the beaver among the Caucasian 
animals ('. ). 

Radde (1866) validated beaver habitats in Upper Svanetia, in the Tskhenis- 
Tskhali Basin, in the vicinity of Lentekhi and in the sources of the River 
Kheledula. Vinogradov (l870) wrote that beavers lived in the bottomland 
of Malyi Zelenchuk on the Trans-Kuban Plain. 

Bogdanov (1873) reported that beavers were found in the rivers of the 
Trans -Kuban steppe, and that the last beaver was killed in 1864 on the 
Laba River, not far from its estuary. Dinnik (1884b) recorded the beaver's 
presence near the source of the Laba. Keppen (1902) published data on 
two beavers killed in 1947 [sic] on the Araks and suggested some interesting 



291 



possibilities of beaver habitation in the upper reaches of the Araks. Linstow 
(1908) published prolific but unsubstantiated data on beavers. 

Satunin (1920) used Keppen's and Linstow's surveys in his history of 
the Caucasian beaver, observing that if the beaver lived in the Supsa Basin 
of Colchis in 1909, it probably survives in some uninvestigated corners of 
eastern Transcaucasia and the northwestern Caucasus. 

Ognev's survey (1947) on the Caucasian beaver with an appended map of 
its former ranges (I ?) added new perplexities for future investigators. 

When studying the swamps and rivulets of the Colchis in 1931 and 1944 
in connection with nutria breeding, the author did not find gnawed tree 
stumps or remains of beaver dams, but was nevertheless convinced that 
beavers could have lived recently in the Rion valley from the source of the 
Kvirila to the Black Sea coast. 

There are rivers (they now have bare banks) near the Tskhaltubo resort 
and in many other places in Colchis which would have been perfectly suitable 
for beaver. 

It is difficult to establish exactly when the beaver disappeared from the 
Caucasus, but it was probably in the middle of the 19th century. 

South of the Caucasus the beaver was known in the past century in 
Mesopotanmia and was widespread in Iran. It inhabited the rivers of the 
Kyzyl-Irmak system in Turkey where an annual yield of 2,000 pelts ('. ) was 
recorded. The beaver might have lived in the swampy area between Kayseri 
and Inesu, south of the middle course of Kyzyl-Irmak in the last half of 
the 19th century (Danford and Alston, 1877, 1880). 

All this data, published first by Keppen (1902) and then by Linstow (1908), 
looks very solid but lacks any scientific confirmation, e.g., figures, 
photos and collections. Radde (1866) provides another example of 
unscientific speculation: he "placed the beaver" on the right tributaries of 
the Rion in the highlands of Svanetia in small mountain fissures with 
turbulent streams avoided even by the otter. 
296 It is difficult to imagine that the beaver could have lived in the central 
parts of Asia Minor and Iran considering the absence of water and forests. 
It is known that beavers feed on wood of poplar, aspen, willow and birch, and 
it is doubtful that dewberry, sea buckthorn, oleaster and smoothleaf elm, 
the main contemporary shrubs and trees growing along the riverbeds in 
Iran and Turkey, would have been sufficient food for the beaver. However, 
the very rapid changes in landscape of the last centuries are a further 
consideration. The recent discovery of beaver remains in the strata of the 
ancient Scythian site of Neapolis in the Crimea showed that fronn the middle 
of the first millennium B.C. beavers lived on the Salgir River on a part 
of the peninsula that was actually dry and completely woodless (Tsalkin, 
1947). With reference to old reports of the beaver on the steppes of the 
southern Ukraine, Pidoplichko (1951) reasonably explained that, given the 
presence of fluvial forests, the beaver could have lived, and could still 
live, in the southern steppe zone, or even in the semidesert zone, on rivers 
crossing these zones. His reasoning is useful in validating the earlier 
distribution of beaver in Ciscaucasia. On the other hand, it is likely that 
the unstable regime of mountain rivers and the dry continental climate 
were limiting factors in the uplands of southwest Asia. 

The beaver cannot live on a rapid, pebble -bedded mountain river for lack 
of shelter and food, even though a forest may grow along its banks. A river 
with a steep gradient provides no place for the beaver to burrow, and every 
flood will inevitably demolish the beaver dams. 

292 



Nevertheless, the beaver survived intensive orogenic disturbances in 
such vast intermontane valleys as the Kura and the Rion, and on submontane 
plains. More frequently, beavers resettled in the area after peneplanation 
had occurred and subsided. 

The history of the beaver on the Caucasus presents a particularly 
interesting phenomenon: the morphological-physiological development in 
an animal native to warm (Mediterranean) climates of characteristics 
adaptive to life under ice. 

These adaptations to life in freezing water bodies have often been 
emphasized by investigators — Linstow (1908), Fedyushin (1935), 
Vereshchagin (1939c) and others — who place their early development in 
the Lower Pliocene with the initial cooling of the climate. Swimming under 
ice and working incisors under water probably involved a considerably 
accelerated process of adaptation in the first half of the Pleistocene. 

Postglacial beaver settlement progressed at a startling rate in territories 
free from inland ice. In view of the difficulty of migration across 
watersheds, the abundance of this animal in historic time in Scandinavia, 
on the Kola Peninsula, and on the Karelian Isthmus might be taken by 
opponents of the theory of a major glaciation on the European plain as 
substantiation for their views. 

The first stage of beaver population decline resulted from excessive 
trading in pelts; the second stage and ultimate disappearance of the species 
from the Caucasus occurred because of anthropogenic alterations of the 
landscape. Environmental conditions changed with the cutting of fluvial 
forests and the acceleration of river currents, erosion and flooding as land 
was cleared and cattle grazed in the basins. These processes were more 
297 rapid along Ciscaucasian rivers than along those of Transcaucasia. The 
beaver survived longest in deep marshes and water courses of the Colchis 
lowland. 

In the 1940's some beavers were brought in pairs from Voronezh into 
the Karayazskii nutria sovkhoz where they reproduced in cages. They could 
be allowed to live out of captivity if they were placed in a suitable environ- 
ment, e.g., riverbanks grown with poplar and willow trees. 

We conclude the following from our survey: 

1. The European beaver inhabited the Caucasian Isthmus in the Pliocene 
and in the Quaternary. 

2. By the beginning of the 18th century, the distribution of this animal 
had become vestigial, surviving only in marshes and rivers of western 
Transcaucasia, in Colchis, and possibly on the piedmont plains of Cis- 
caucasia along tributaries of the Terek and the Sunzha. The beaver is not 
known in mountain regions of the Caucasus in historic time. 

3. The final extinction of beaver on the Caucasus occurred in the middle 
of the 19th or the beginning of the 20th century in Colchis. 

4. If water and forest regimes and hunting can be regulated in the 
U.S.S.R., it will become possible to breed beavers in small numbers on 
some of the left tributaries of the Kuban and on the rivulets of the Khachmas 
lowland and of Colchis. Especially suitable for breeding grounds are those 
spring-type rivulets ("kara-su") with a constant flow of water and plantings 
of poplar and smoothleaf elm on their banks. Beaver breeding, however, 
will not produce any marked effect on an accelerated agricultural economy. 



293 



Family МиКШАЕ 

Remains of 15 species of this family are known from Quaternary deposits 
on the Caucasian Isthmus; today the Muridae of the Caucasus comprise 
29 species. The emergence of the genus Rattus is of particular interest. 
Black rat — R at t u s rattus L. R. cf. rattus remains were known 
long ago from Quaternary strata of numerous Western European caves from 
the British Isles to the Carpathians (Woldrich, 1882; Brandt and Woldrich, 
1887). They were usually found with bones of animals of the Upper 
Pleistocene — cave bear, woolly rhinoceros, and others — in Upper 
Paleolithic strata, such as Aurignacian strata of the Evelinis cave of 
Somerset (England). The majority of black rat bones were found along 
Italian coasts and generally around the Mediterranean (Wolf, 1939). 

In Lebanese, Syrian and' Palestinian caves black rat bones were found 
in the Mousterian strata in which were also found remains of Rhinoceros 
mercki and the hippopotamus (1) (Bate, 1937). Black rat remains are, as 
a rule, deposited contemporaneously with their enclosing strata, i.e. , 
"in situ, " but there are probably many cases of contemporary rat remains 
being thrown down into the bottom of excavations or of the rats themselves 
entering the deeper strata by fissures. 

Nevertheless, the explanations are insufficient to accept the theory of 
postglacial, prehistoric penetration of the black rat by "importation" into 
Europe — a theory perpetuated in scientific articles. The black rat is 
probably a Pliocene relict of the Mediterranean. It survived the Pleistocene 
on the warm coasts of southern Europe and north Africa, and perhaps even 
on the Caucasus (Vereshchagin, 1947d, 1949c). 
298 In the U.S.S.R. black rat bones were found in Greek and Roman strata 
of Olvia, in the southern Ukraine (Pidoplichko, 1938c). 

On the Caucasus single bones of 
this kind were found in Neolithic strata 
of the Akhshtyrskaya cave and in 
graves with jar burials of the first 
millennium B.C. on the Kura near 
Mingechaur (Figure 122). 

Actually the black rat is widespread 
on the Caucasus, living both "wild" and 
as commensals with humans. It lives 
within the harbor constructions along 
FIGURE 122. Black rat pelvis from strata of the the Azov, Black and Caspian sea 
first millennium B.C. in Mingechaur coasts. The black rat is not found in 

the mountains or plains of Ciscaucasia 
and probably has never lived there. 
Black rat bones were not observed among the thousands of bones 
recovered from eagle owl pellets which were collected by Kistyakovskii 
(1935) and the author from the Stavropol and Pyatigor'e areas. 

Black rat is widespread along the Black Sea coast from Batumi to 
Novorossiisk, living in gardens and forests, and infrequently on the banks 
of water bodies where it has been largely replaced by the Norway rat. 
Flerov (1927) observed it north of Batumi forest. Shidlovskii (1947, 1948, 
1950) recorded the rat in many parts of Abkhazia, Mingrelia and Adzharia. 
He claimed that the black rat can be found in Abkhazia from sea level to 
600 m and in Adzharia to an altitude of 1,500 m. The author found both 



294 




the black rat and the Norway rat on Bebesyri Lake in the fall of 1946, but 
it has hever been observed in the bogs around the Paleostom Lake 
(Vereshchagin, 1941b). According to Stepanov (l93l), the black rat of 
Batumi is less numerous than the Norway rat. In eastern Transcaucasia 
it lives "wild" in the forests of northern Armenia where it is also found in 
populated areas. In 1937 it was frequently observed in the Shamlug mine 
and in the mills of the villages of Kulali, Achadzhur and Marts (Shidlovskii, 
1942). The animal is rare in South Ossetia and is found only in inhabited 
places deep in ravines . In Kakhetia and the Alazan- Agrichai valley it lives in 
gardens and buildings, quite often in garrets (Argiropulo, 1940a). It is 
common in summer in blackberry thickets bordering fields near Kutkashen 
and Vartashen. In the Kura lowland it lives in oases, especially in black- 
berry thickets along irrigation ditches. 

In Baku the black rat finds refuge in structures along the littoral strip 
and usually is not found farther than 100 m from the bay's edge. Sixty 
percent of the yield of a large-scale capture was from ships in Baku port 
and only 9% from port buildings. The remainder of the yield was made 
up of the Norway rat (Vereshchagin, 1942a). 

This rat is common in the Lenkoran lowland from the village of Masally 
to Astara, and further along the southern coast of the Caspian sea in Gilan 
and Mazanderan. In these areas it lives in orchards and can be found in 
all harbors, especially those near fisheries. Rodents from Asterabad were 
known to zoologists even at the beginning of the last century (Menetries, 
1832; Radde, 1886). 

According to records from large-scale catches in Lenkoran and adjacent 
villages (late 1930's-early 1940's), the black rat comprised only 0.51% 
299 of all synanthropic rodents caught (Vereshchagin, 1949d). Even here the 
rodent does not inhabit m,ountains but confines itself to lower altitudes. 
The ability of the black rat to exist in the wild the year round in Colchis 
and Asterabad and to construct burrows under tree stumps, fences and 
shelters in hollows substantiates its Pliocene origin in the Mediterranean 
area in general, and in Transcaucasia in particular. 

The black rat adapted readily to the new types of shelter provided by 
human culture: first mud huts, and later brick, stone and wood buildings 
and boats and ships, in which heat and food were available. Adaptation to 
life within harbor constructions and on ships is, in our opinion, associated 
with the rat's original habitat on the warm Mediterranean coasts, where 
feeding along the shore was possible. Later, penetrating inland along rivers 
and confining itself to inhabited places, it gradually occupied the lower 
Transcaucasian forests. 

Norway rat — R a 1 1 u s norvegicus Berk, There are no proved 
remains of this species from the Pleistocene strata of southwest Asia, the 
Caucasus and Western Europe. Remains of Norway rat found in the caves 
of Czechoslovakia, Malta, Gibraltar and Ireland were bedded together with 
the remains of postglacial animals and the type of preservation does not 
indicate an early origin (Wolf, 1939). The bones of Norway rat are known 
on the Caucasus from later Middle Ages strata (l5th-16th century, A.D.) 
of the excavation of the courtyard of the palace of the Shirvan Shahs in Baku 
(Vereshchagin, 1949c), and from eagle owl pellets, no more than several 
hundred years old, found in small caves in the Stavropol and Pyatigor'e 
areas and in North Ossetia and Imeretia. 



295 



The first reference to mice as being harmful to crops and trees was 
made by Claudius Aeiianus in the 3rd century A.D. (Latyshev, 1904). 
Arabs probably knew of the Norway rat in the early Middle Ages (Zhitkov, 
1944). 

Nowadays the Norway rat is reported on the Caucasus from all inhabited 
places of the country which are connected by railways or roads, from 
coastal villages, fisheries and ports along the shores of the Caspian and 
Black seas, and from the rivers (Kuban, Sulak, Terek, Kura and Rion), 
especially from their navigable sections (V'ershchagin, 1947d) (Map 52). 

On the Caucasus the Norway rat is frequently found living off the land, 
especially in the humid foothills of Kabarda, North Ossetia and the Grozny 
Region. Here the Norway rat is often a real pest of gardens and field 
storehouses ("sapetki") (Rossikov, 1887; Bogdanov, 1936). In Kabarda and 
North Ossetia the Norway rat lives in mountain villages ("auls"), 
particularly in the vicinity of mills, almost to the upper areas of the 
Greater Caucasus ravines (Turov, 1926c). 

There are wild populations of this rat in the Colchis swamps (Figure 123^ 
in Asterabad and on ancient lakes of the Kura left bank in Azerbaidzhan 
(Vereshchagin, 1914b, 1942b, 1947d, 1949d; L'vov, 1949). 

The Norway rat also lives in rock placers on the Caspian desert coast, 
e.g. , on the Apsheron Peninsula, where it feeds on flotsam. In the Rion 
and Kura lowland swamps, the animal shelters in open, raised nests in 
reed thickets or in closed shelters constructed beneath alder roots. 

In the wild, the Norway rat feeds primarily on vegetable food but its 
consumption of animal food is also significant. 




FIGURE 123. Habitat of Norway rat in Colchis 

Photograph by author, 1939 



296 



During periods of prolonged snow cover and hard-frozen open waters, 
some of the Norway rat population of reed and alder swamps in Trans- 
caucasia die in their own habitats, and some survive in the inhabited areas 
along the shores (Vereshchagin and Dyunin, 1949). 

This species, closely associated as it is with water biotopes, probably 
could not tolerate glaciation either in Ciscaucasia or Transcaucasia. 

The wider contemporary distribution of Norway rat throughout the world, 
as compared with black rat, is consonant with its Pliocene origin, and 
with the Mediterranean Holocene origin of the black rat. 

The Norway rat possesses a greater adaptability of physiological functions 
than the black rat (Ruttenburg, 1950) which has facilitated its settlement 
along with man's from the tropics to the Arctic. However, it is less 
resistant to winter conditions in the wild in Transcaucasia because of its 
hydrophilic nature. 

In developing Zhitkov's (1944) opinions and our own main premise of 
the survival of the genus Rattus in the Mediterranean from the Pliocene 
(Vereshchagin, 1949c), it was necessary to consider the Norway rat as a 
postglacial newcomer into the Transcaucasian swamp fauna and to the 
forests of the Ciscaucasian lower mountain zone. This raised questions as 
to the routes and dates of its penetration into the Caucasus. 

The Norway rat certainly could not have entered Europe from central 
Asian deserts. The more recently popularized notion that the species 
entered the Caucasus by way of caravan routes across the dry Iranian 
desert from India was properly discredited by Kashchenko in 1912. The 
301 route of penetration obviously was along the sea coasts. In view of the 
historical fact that Egyptian and Phoenician ships crossed the Red and 
Mediterranean seas as early as the second millennium B.C. , later sailing 
as far as southern Asia (Avdiev, 1953), it seems clear that the Norway rat 
was transported into the Mediterranean before the common era, probably 
first to the coast of the Black Sea in the Caucasus. 

The penetration of the Caspian coast occurred later, after settlement 
of the Black Sea area, either by water — on the Manych — or by land — 
very possibly along the humid, sloped plains of Ciscaucasia. 

Before urban concentrations of many-storied buildings and slaughter 
houses, and the development of canal, railroad and automobile trans- 
portation, the Norway rat apparently lived wild in reed-grown swamps and 
river deltas, as can be seen from observation of its habitats in the estuaries and 
lower reaches of the Danube and the Dniester (Aizenshtadt, 1950b), in the 
swamps of the Rion, and in the Volga delta. 

Pallas [1741-1811] observed rats in the Volga delta; at that time, the 
Norway rat probably lived there, as did the water vole. 

The new developmental stage of the Norway rat range in the Caucasus is 
associated with the construction of paved roads and railroads, and rural and 
urban centers in the 19th and 20th centuries. Slowly it extended into the 
populated areas of the Kura lowland along waterways (the Kura and Araks 
rivers and their tributaries). The rats first penetrated the Alazan-Avtaran 
valley by the left tributaries of the Kura: the Turyan-Chai, Geok-Chai and 
Alindzha-Chai. Later the population growth was accelerated by road 
construction — less so on the dry Armenian plateaus than elsewhere. In 
Yerevan and Dzhulfa, for instance, the Norway rat appeared in the 1930's after 
the construction of the railroad connecting these towns with Tiflis and Baku. 



297 



In the Ciscaucasian Plain the settlement of the Norway rat accelerated 
in the 1920's with the development of road transport. Gubarev (1941) 
recorded the appearance of the Norway rat in inhabited areas of the Salsk 
steppes. 

The Norway rat was known to Satunin (1901) in Stavropol from bones 
contained in eagle owl pellets from Mount Strizhament, which established 
that the species inhabited the forested valleys of this region as early as 
the last century. 

Kuzyakin's survey (1951) on the Norway rat's origin in northern Eurasia 
and its relationship to man added nothing to an understanding of the history 
of the animal in the U. S. S. R. 

Striped field mouse — A p о d e m u s agrarius Pall. Remains of the 
striped field mouse are known from Pleistocene strata of Czechoslovakian 
caves (Wolf, 1939). They are as yet unknown on the Caucasus, although 
they have been exposed in Upper Quaternary deposits of the lower Don. 

The contemporary range of the striped field mouse occupies the broadleaf 
forest zone and the southern Eurasian taiga with discontinuities in Trans- 
baikal and in the upper Amur Basin (Bobrinskii, Kuznetsov and Kuzyakin, 
1944). The striped field mouse inhabits only the river valleys of the Russian 
Plain steppe zone. It is now absent from the Crimea although it has been 
found in Pleistocene strata. The distribution in Ciscaucasia, now isolated 
from the north, is confined to the humid piedmont and to the riverbeds of 
o^2 the steppe and semidesert (Sviridenko, 1944). Striped field mouse has 

been reported from the Black Sea coast south as far as Sukhumi (Shidlovskii, 
1947, 1950). 

The species inhabits the Stavropol Plateau and the tugai forests of the 
Kuma, Terek and Sulak rivers. An isolated range is situated to the southeast 
in the forests of the Khachmas lowlands from the lower reaches of the Samur 
to Khachmas (Beme, 1928; Vereshchagin, 1944). 

Its vertical distribution is insignificant: the upper parts of inhabited 
mesophytic meadows and fields in the northwest Caucasus between altitudes 
of 950-1,000 m; in the Nalchik area, to an altitude of 700 m; and near 
Ordzhonikidze and in Alagir, between 800 and 850 m. 

Sviridenko (1944) observed that striped field mouse distribution is related 
to the amount of precipitation. The animal becomes eurytopic only if the 
annual precipitation is above 500 mm. In more arid regions the rodent 
inhabits only humid biotopes like the lowland forests of the Khachmas 
depression near the Caspian coast. Here the striped field mouse lives in 
the humid bottoms of afforested valleys and on the banks of spring-fed brooks 
under cover of alder and blackberry growth interspersed with liana. In this 
environment the mouse is not harmful to cereal crops as it is in other places. 
On the more arid Kusary sloping plain and on the remainder of the Caspian 
coast, this animal is not encountered. 

The distribution of striped field mouse in Ciscaucasia is mostly confined 
to regions with a dense network of small torrents and springs, such as the 
developmental zone of piedmont trains. Shrub areas along small rivers 
and small meadows along turbid brooks are characteristic biotopes. 

Consequently, the Caucasian foothills contain several isolated ranges: 
the Kuban (including the Black Sea coast), the Terek-Sunzha and the Samur 
(Map 54). The breaches in the Ciscaucasian range areas are created by 
dry, steppelike divides. 



298 



Earlier — in the Pleistocene — the striped field mouse barely penetrated 
farther south than the lower forests of Khachmas, since its remains were 
not found in the Pleistocene asphalts of the Apsheron Peninsula. This is 
further substantiated by its absence from the Alazan-Agrichai valley where 
conditions are suitable for its habitation. 

In humid western Transcaucasia the striped field mouse is distributed 
along terraces of the Black Sea coast southward to Sukhumi. Here it is 
frequently observed in glades, clearings, shrubs and orchards. 

Shidlovskii's data (1947) (Table 70) are characteristic for striped field 
mouse distribution in the biotopes on the Black Sea coast. 

TABLE 70. Distribution of the striped field mouse according to biotopes in Abkhazia 









u 




с 


-а 






э 








1) 

JJO 

-a 


Я с 
> oo 


и 

M 

О 


о 


В 
о 

и 


"о о 

-л > 


X) 

2 

л: 
ел 


3 

.S а 

5 5 


2 
о 
Н 


Nmnber of 


(absolute 




















animals > 


number 


1 


27 


3G 


12 


10 


31 


135 


2 


254 


caught J 


in %) . . 


0.4 


10. fl 


14.1 


4.7 


3.9 


12.2 


53.2 


0.8 


100.0 



303 These data show that anthropogenic alterations of the topography in the 
Black Sea littoral zone, especially deforestation, favor the growth of the 
distribution and population of the rodent. It is possible that the considerable 
extension of the striped field mouse range south along the Black Sea coast 
in historic time was promoted by man-made changes in the landscape. 
However, farther south in Colchis, important to note, there are no 
occurrences of striped field mouse, although the biotopes are nnuch like 
those of Abkhazia, a fact probably explained by the excessive hunnidity. 

The peculiar underdevelopment of the field mouse ranges on the Caucasus 
suggests that the penetration of this species from the north occurred rather 
late in the Pleistocene. With the postglacial advance, the Caucasian 
populations of the striped field mouse were apparently cut back by the 
development of a large belt of dry steppes. 

The absence of the animal from the Crimea in the Recent and in the 
Pleistocene also confirms a late development of the range in the south. 

In the Recent epoch, the striped field mouse's distribution on the Caucasus 
increased in the southwest with forest clearing in places of excessive 
moisture, and in the dry regions to the north and south with the developnnent 
of irrigation. 

Asia Minor hamster — M e s о с r i с e t u s auratus Water, (s. lato). The 
question of the origin of the hamster of genus Mesocricetus on the 
Caucasus is rather complicated, for the earliest hamster remains — 
genus Paleocricetus — found on the Caucasus date from the Middle 
Miocene. In the Lower Pliocene of Ciscaucasia there lived a hamster 
almost the size of the golden hamster of the eastern Mediterranean area. 
Thus, the Caucasus could have been as much of a focus of Asia Minor 
(golden) hamster development as was southwest Asia. 



299 



Fragments of hamster remains of the genus Mesocricetus are known 
from the Middle and Upper Paleolithic of Palestine (Athlit cave; Picard, 
1937) and, within the limits of the Caucasus, from the caves of the Rion 
Basin and Kvirila River (Gvardzhilas, Mgvimevi, Kudaro) and from 
Pleistocene strata of the Apsheron Peninsula. 

It is important to emphasize that, until now, the only hamster remains 
found in Quaternary deposits of the Russian Plain have been those of the 
genus С ricetus. 

Postglacial cave deposits of Asia Minor hamster remains are quite common 
on the Armenian Highland, in central Transcaucasia and in central 
Ciscaucasia. 

In recognition of its considerable geographic variability, the Caucasian 
representatives of the genus Mesocricetus are classified by 
taxonomists under two or three separate species. Accepting southern 
southwest Asian origin of the extant Caucasian golden hamster, we can 
further assume that, in chronological order, Mesocricetus, 
M. auratus raddei Nehr. , and M. auratus nigriculus Nehr. 
comprise a genetic series of successive subspecies. 

The contemporary range of this hamster includes Syria, Asia Minor, 
the Armenian Highland, Talysh, the northwestern Iranian Plateau, eastern 
Transcaucasia, central Dagestan and the Ciscaucasian Plain. Surveys of 
the distribution areas were given by Argiropulo (1935, 1939a), Neuhauser 
(1936), Shidlovskii (1940c) and Ellermann (1948). The vertical distribution 
of Mesocricetus lies within a range of 0-2,400 m. The altitudinal 
optimum of the Transcaucasian subspecies M. auratus koenigi Sat. 
is, according to Shidlovskii (1940c), in the range of 1,200-2,000 m (Map 58). 
304 On the Armenian Highland within the limits of the U. S. S. R. , this 

hamster is widespread in relatively warm and arid upper river valleys. 

It inhabits the festuca-grown upland steppe of the Pambak River valley, 
the borders of Lake Sevan, and the Saraibulakh and Zangezur ridges. On 
the cold upland of Dzhavakhetia it can be observed on the steppe meadows 
near Tabistskhuri Lake. 

In the Karabakh uplands the hamster is known from the Gevorkevan 
vicinity. 

In Talysh it has been encountered in an area stretching from the upper 
reaches of the Vilyash-Chai south to Kel'vyaz, in the upland zone of 
artemisia— grass -steppe at an altitude of 1,600 to 2,000 m. It was not found, 
however, in groupings of small goat's thorn and acanthus as in the 
Armenian Highland but, rather, concentrated in more humid places in the 
valley bottoms, especially on the margins of sown fields. 

The animal's distribution on the ranges of the northern Iranian Plateau 
is probably discontinuous. 

In eastern Transcaucasia the hamster inhabits the Gori depression, 
occupying regions of earlier steppes or forest clearings. In the Pleistocene 
and Holocene, it was distributed in Imeretia and in the upper reaches of 
the Rion. Eastwards, on the lora and Kartalinia plateaus, the animal 
inhabits areas grown with the racemose andropogon at altitudes of 
550-600 m. It is not found east of the Alazan to Shemakha, but it is possible 
that relict colonies will be found northeast of Shemakha, for example, in 
the Kilyazi-Chai and Gil'gin-Chai valleys, and west of Derbent. 



300 



The distribution of the Dagestan subspecies, M. auratus raddei 
Nehr., is limited in inner Dagestan by the valleys of the Argun, the 
upper Sulak and the Samur. The animal has been traced from the village 
of Shatili to Evdokimovskoe in the Argun valley. According to Rossikov 
(1887) it is known from the Andi Koisu River. Accordingto the data from the 
Orlov expedition, Beme's records (1925), and Dyukov (1927), the hamster 
is encountered on the whole of the Khunzhakh uplands on the fields 
surrounding Baitl, Chondotl and other villages. 

Here the hamster inhabits the mountain xerophytic steppe and is attracted 
to fields of grain. Its vertical placement is from 1,600 to 2,300 m (Geptner 
and Formozov, 1941). 

In the Middle Pleistocene these ranges were probably linked with the 
piedmont areas of the Apsheron. 

Perhaps this link was broken in the Upper Pleistocene when Dagestan 
was divided by water erosion and the formation of a forest zone on its 
southeastern border. 

The present-day range has enlarged somewhat under the influence of 
agricultural activity. 

The distribution of the Ciscaucasian form M. auratus nigriculus 
Nehr. is very extensive, although always "squeezed" close to the mountains 
(see Map 58). 

Its habitat in northern Ciscaucasia is in the zone of artemisia— festuca- 
steppes, while on the Stavropol Plateau and in the Kabarda foothills it is 
found on steppe-meadows and plots planted to corn. The animal is also 
encountered in the motley-grass of the Terek valley (Ognev, 1924). 

In many places this hamster has actively penetrated the mesophytic 
foothills of Ciscaucasia where the forests have been cut, for instance on 
the Trans -Kuban, Kabarda and Terek-Sunzha sloping plains. In general, this 
Ciscaucasian form is more eurytopic than its Transcaucasian parent, 
although, in the main, it occupies dry places. 
305 Its vertical distribution in Ciscaucasia normally extends from sea level 
to 1,000 m. In central Ciscaucasia, it penetrates the mountains as far as 
Kislovodsk andNal'chik, but is not found in dry, longitudinal valleys of the 
northern slope. It is spread along the valleys of the left tributaries of the 
Terek. M. auratus nigriculus is distributed only as far as the 
Chernye Gory canyons. 

In the last few decades the settlement of hamsters in the Rostov Region 
has been reported and in the 1940's they were found northwest of Manych. 

Judging from the Asia Minor hamster's contemporary range and from its 
level of trophic succession, it seems possible that the ancestral species 
penetrated the Caucasus from the south and southwest during the Tertiary, 
most probably during a dry phase of the Upper Pliocene. 

The path by which the hamster extended its range from the Lesser 
Caucasus to eastern Ciscaucasia probably passed the foothill steppe zone 
along the base of the southeastern slope of the Greater Caucasus and farther 
through the present uplands of Dagestan which at that time were lowlands. 
Later this route was impeded by loam and pebble sediment on the swamped 
plains of the piedmont, particularly in the region of presently eroded 
sections of Adzhinour and of the Kusary sloping plain. The distribution area 
of the hamster was probably the widest in the Middle Pleistocene during 
the wide steppe formation on mountain slopes. The subsequent mesophjrtic 
trend in landscape and the shifts in altitudinal zones during glaciation 



301 



created a break and caused the decrease of the hamster range. Another 
breach in the range occurred with the onset of the postglacial xerothermic 
period which produced a discontinuous upland mountain relief in southwest 
Asia. 

On the other hand, in the humid mountain regions of Imeretia and South 
Ossetia the Lower Holocene desiccation of ridges and slopes contributed 
somewhat to the spreading of the hamster range; there it was only displaced 
later during another humidification of the climate. 

There can be another interpretation of the history of the hamster range 
development on the Greater Caucasus and in Ciscaucasia, if, while accepting 
two distinct species for the Dagestan and Ciscaucasian forms, we consider 
the possibility of their having a common ancestor in a local Tertiary form. 

In the Recent epoch, ranges preserved in southern Transcaucasia are 
relatively stable and are diminished only during cattle grazing, large 
scale plowing of artificially watered land, or purposeful killing of the 
animal as a pest through the use of chemical exterminators. 

Asia Minor gerbil — M e r i о n e s blackleri Thos. Remains of gerbils 
of the genus Me r tones were recorded from Acheulean and Mousterian strata 
of Palestinian caves (Picard, 1937; Wolf, 1939). In the Caucasus, aside 
from findings of Gerbillus sp. in Pliocene strata of the Armenian 
Highland (Bogachev, 1938c), only the remains of the present-day species 
are known from the superficial layers in caves and shelters of the Araks 
valley. 

The present-day range of Meriones blackleri Thos. includes 
Asia Minor, Syria, western Iran and a part of Transcaucasia (EUermann, 
1948; Bobrinskii, Kuznetsov, Kuzyakin, 1944; Pogosyan, 1949). 

In Transcaucasia Meriones inhabits the middle Araks valley, certain 
places in the Araks gorge in the Zangezur ridge, the eastern Karabakh 
foothills, the Mil'skaya steppe, the southern Mugan steppe and the Kirovabad 
sloping plain. In the wide valley of Akera-Chai, it is found almost as far 
as Lachin, that is, about 1,500 m above sea level. The population of 
306 Meriones blackleri Thos. in the plains of the middle Araks Valley 
is greater than that of other species; on the stony slopes, however, it is 
superseded by Meriones persicus Blanf. The animal is not encountered 
on the left bank of the Kura, although conditions are the same there 
as on the right bank. On the Kirovabad sloping plain its distribution can 
be traced to the lower reaches of the Khram. According to data from large- 
scale catches, it has a greater population than Meriones erythrourus 
Gray in the foothills of the Lesser Caucasus. 

In the Mil'skaya steppe M. blackleri can be seen with M. ery- 
throurus almost everywhere, but in the Karabakh foothills it is 
encountered only in warm areas with artemisia groupings at altitudes of 
350-400 m above sea level. In the northern part of the Mugan steppe, this 
animal has been largely dislodged by cultivation and irrigation and is seen 
mainly in artemisia semidesert along the border from Karadonly to 
Belyasuvar. On the northeastern slopes of Talysh, it has been traced to 
the Eshakchi outpost at altitudes of about 600-700 m (Map 63). 

The animal mainly inhabits alluvial and talus deposits of loessial 
loam in large fluvial valleys grown with artemisia and ephemeral grasses. 

In view of M. black leri's specific distribution in eastern Trans- 
caucasia and, more particularly, its absence from the Kura-Iora interfluvial 



302 



area and the Shirvan steppe, we can assume that it settled in the north only 
relatively recently, i.e. , in the Pleistocene or perhaps even in the 
Holocene. In southern Transcaucasia in the middle Araks valley, 
M. blackleri may be considered a local species from the Upper Tertiary. 

Persian gerbil — Meriones persicus Blanf . Fossil remains of 
this species from Pleistocene strata have not been validated, but they are 
known from Holocene cave deposits on the southern slopes of the Armenian 
Highland and from the middle Araks valley. 

The main part of the M. persicus range (Map 62) includes Turkey, 
Iran, Afghanistan and southern Turkmenia (Neuhauser, 1936; Geptner, 
1940; Ellerman and Morrison-Scott, 1951). 

On the Caucasus it is eurychoric over the southern slopes of the 
Armenian Highland and in the highlands of Talysh. Its population in the 
Alindzha-Chai Basin, according to large-scale catches, comprised 62.5% 
of the total Meriones population. It is not encountered on the left bank 
of the Araks gorge north of Mindzhevan. In the gorges of the southern 
slopes of the Zangezur ridge, the range reaches an altitude of 1,800 m in 
astragali— tragacanth formations and stretches as far as Zangelan and 
Shakhbuz. 

On the Saraibulakh ridge, according to Dal' (1904b), Meriones 
comprised 17% of the rodent constituent of the food of the eagle owl. It is 
observed in the Zanga valley to an altitude of 1,230 m (Dal', 1948b) and 
is common on the slopes of the outlying northern ridges of the middle Araks 
valley among mountain halophytes of the semidesert (Flerov and Gureev, 
1934; Argiropulo, 1939b; author's data, 1947). 

In the highland steppe of Talysh the species is comimon in the upper 
reaches of the Vilyash-Chai and Zuvanda at altitudes of 1,500-1,800 m. 
Our data collected in the vicinity of the Dyman outpost in 1945 show that 
its remains constituted 26 % of the remains of all rodents found in owl 
pellets. 

In the dry mountain areas of northern Iran and Asia Minor M. persicus 
is evidently the most numerous and eurychoric. Its usual biotopes are the 
stony slopes of valleys which are dotted by cushions of thorny astragali, 
307 acantholimon and artemisia, and in spring by ephemeral grasses and 
poppies. In certain places the animal emerges on divides beneath the 
mountain-grass and steppe zone. 

The facility of M. persicus in digging its burrows under stones or 
shrub roots, its skill in climbing stones, its hairless feet, long tail with 
panniculus and its long vibrissae — all these features, according to 
Argiropulo (1939a), are early adaptations to a mountain desert habitat. 

This animal is mainly granivorous: during early summer, it feeds 
mostly on grain of small grasses; in later summer, on grain of mountain 
saltwort. It is an early indicator species of the southwest Asia mountain 
desert mammal complex. 

M. persicus thrives in agricultural regions. In grain fields and along 
bordering fences, its population is always increased somewhat. Cattle 
driving and overgrazing contribute to the spreading of thorny astragali, thus 
extending the range. 

Red-tailed bibyan gerbil — M e r i о n e s erythrourus Gray(s. lato). 
Fossils of this animal are known from Middle Pleistocene strata of the 
Apsheron Peninsula. 



303 



At present, this polymorphous species (for which as many as 23 
subspecies have been recorded) is identified with M. libycus Licht. Its 
vast range includes north Africa, southwest Asia, eastern Transcaucasia 
and central Asia to Semireche [Dzhety-Su]. Its northern limit on the 
Caucasus passes the latitude of Shemakha (Ellermann and Morrison-Scott, 
1951; Bobrinskii, Kuznetsov, Kuzyakin, 1944). 

In Transcaucasia it inhabits all of the Kura-Araks lowland and foothills — 
from Tiflis on the west to the Caspian coast on the east (Map 61) — and 
is especially numerous on the Apsheron Peninsula. M. erythrourus has 
not been observed north to Kilyazi or south to the Kyzyl-Agach Gulf on the 
Caspian coast. 

The species is encountered throughout the central lowland. A dense 
•population is noted among artemisia, saltwort and caper formations in the 
flat semidesert of the Mil'skaya, Shirvan and Mugan plains. The animal 
settles more frequently on alluvial ranges and on the margins of minor 
solonchak depressions overgrown with arboreal saltwort shrubs and 
Austrian and Sovich wormwood. The maximum population is observed, 
however, on the sandbank ridges of the Kura, Alazan, lora and Araks, 
especially in the tamarisk-thicket zone. Similar high-density populations 
are encountered on the dunes of the Caspian coast where the sand hillocks 
are held by tamarisk and ephedra. In such places there are perhaps 
40 specimens per hectare. Biotopes created by man, especially vineyards, 
abandoned irrigation systems, shepherds' camps in the steppes and fallows 
of several years standing, are of great importance to this animal's settlement. 

The animal is crepuscular and nocturnal in sumnner, crepuscular and 
diurnal in spring and fall, and diurnal in winter. Meriones erythrourus 
is mainly granivorous: in spring it feeds on shoots of ephemeral annual 
plants; in summer, on shoots and grains of goatgrass, ragweed, brome 
308 and insects, in particular on non-gregarious locusts; in fall, on new shoots 
of ephemeral grasses; in winter, on wormwood stalks, seeds of arboreal 
saltwort and others. 

In its search for grain, the animal climbs high on the saltwort shrubs. 
During snowfalls these animals usually lie in burrows untilthe thaw, rarely 
grazing on adjacent shrubs. The abundance of animals on sand hillock 
ridges in level steppes is explained by the ease of constructing ventilated 
burrows and the opportunity of winter fattening on artemisia and saltwort. 
The coincidence sometimes seen of dense rodent populations in cattle camps 
("yatagi") in these areas might persuade an unskilled observer that cattle 
raising is favorable to hibernating populations of Meriones (Gladkina, 
1952). 

The contemporary range of M. erythrourus and its adaptation 
to the landscape indicate that the species formed in the southern and 
southeastern Mediterranean range. It penetrated the Caucasian Isthmus 
from the south and southeast probably as early as the Pliocene but no later 
than the Lower Pleistocene, since by the Middle Pleistocene it already 
inhabited the Apsheron Peninsula. 

The animal survived the transgressions of the Caspian Quaternary on 
the Tertiary hillocks on the northern and western borders of the Kura Bay. 
In the eastern part of the Kura-Araks lowland, this species is one of several 
late settlers of the territory after the sea waters retreated from it in the 
Holocene. 



304 



Local agriculture has, until now, been favorable for the development of 
the species. However, the construction of irrigation works and the expansion 
of irrigated cotton fields will rapidly drive this animal from vast land areas. 

Steppe vole — M i с r о t u s socialis Pall. (s. lato). Remains of this 
genus have been found in Middle Pleistocene strata of the Apsheron 
Peninsula, in Holocene cave strata of the Saraibulakh Range in Armenia 
and in many other places on the present-day range, which includes 
Cyrenaica, the Balkan Peninsula, the southern Ukraine, the Crimean steppes, 
southwest Asia, central Asia, eastern Ciscaucasia and eastern Transcaucasia 
(Minin, 1938; Vinogradov and Argiropulo, 1941; Ellermann, 1948). 

In Ciscaucasia this vole inhabits the Salsk steppes along the Manych 
valley, the regions of the lower Malka and Baksan, the slopes of the Terek 
and Sunzha ranges and steppelike piedmonts of northern Dagestan. The 
animal penetrated the Terek-Kuma massif of the sandy semidesert along 
the network of ducts and channels of the Kuma, Kura and Terek. 

On the sloping plain of the Kusary, the steppe vole's range is found in 
warm places to an altitude of 600 m, adjacent to that of the common vole. 
The animal inhabits only wide plowed fields in lowland forests of the 
Khachmas plain. 

From the western Caspian coast the range extends into eastern 
Transcaucasia at altitudes of 700-800 m on piedmont ridges (Map 72). 

The vole penetrated the lowland forests of the Alazan-Agrichai valley 
through deforested areas. On the Apsheron Peninsula and in Kabristan, 
there are small relict colonies in places with a maximum annual 
precipitation of only 280 mm, relieved, however, by localized, supplemen- 
tary moistening of the soil and an underlay of impervious layers of 
Apsheron limestone. The continuous distribution of the animal in the Kura- 
3 09 Araks lowlands begins in a region where the annual precipitation is less 
than 350 mm. In southeastern Mugan the distribution toward the Lenkoran 
lowlands is limited by greater soil moisture in winter and 800 mm annual 
precipitation (Vereshchagin, 1942c, 1946b). ■•' 

In the Kura valley the vole was found only as far as the Mtskheti until 
recently, when it penetrated the Mukhran valley as a consequence of 
deforestation there. Near the Mtskheti the range is broken by the forested 
narrow gorge of the Kura, and resumes again in the dry Gori depression 
at altitudes of 600-800 m. Here, M. socialis goriensis Arg. was 
isolated as a subspecies, slightly different from the typical species. The 
vole inhabits the Kirovabad sloping plain up to the existing lower timberbelt, 
i.e. , to 600-700 m, although in some gorges it is encountered even higher. 
On the deforested northeastern slopes of Karabakh, the steppe vole has 
spread upward to 800 m in the vicinity of Madagiz, Mardakert, Martuni 
and other points where its range is adjacent to the lower distribution belt 
of the pine vole. The vole penetrated high into the mountains through the 
valleys of the Akera and Okhchi-Chai via the southern slopes of Karabakh. 
A similar penetration can be observed in gorges of the southern slope of 
the Armenian Highland. 

In the semidesert of the Kura-Araks lowlands the steppe vole forms 
temporary settlements in caper formations on chestnut soils, and lesser 
settlements in the artemisia— grass groupings. Summer conditions for the 
animal in the semidesert zone are rather poor because of lack of moisture 

K.N. Rossikov's report on the mass daytime migration of voles during August in the Mil'skaya Steppe, 
noted by Ognev( 1950) is, in fact, only a fiction (Vereshchagin, 1946b). 



305 



in food and the high temperature of the soil. The vole withstands these 
conditions by feeding partially on locusts and mollusks and by constructing 
deep burrows (Vereshchagin, 1946b). 

The semidesert population of voles declines catastrophically each year 
as June approaches because of cessation of reproduction and the sudden 
death of mature animals brought about by the desiccation of food and the 
high temperatures. Viable isolated colonies of animals survive in some 
hollows, in ravines grown with milk thistle (Silybum), in gardens 
moistened by irrigation, or under caper brush, etc. 

In the Mugan steppe, in the area of Belyasuvar and Astrakhan-Bazar, 
the years of peak breeding witness haphazard nocturnal mass shifts of 
the steppe vole in the first half of June because of physiologic disturbances 
aroused by the desiccation of vegetable food and soil overheating (Mamedov, 
1950). * 

Because vole behavior shows seasonal peculiarities controlled by 
temperature and insolation, in Azerbaidzhan this animal is more vulnerable 
to attack by predatory birds and mammals during the winter. 

Characteristic "pulsations" in the ranges of this species can be traced 
over the years near the base of the Apsheron Peninsula. When winters are 
warm and humid and summers cool for several consecutive years, there 
is an acceleration in reproductive activity and a significant increase in 
population. In the semidesert when the fall is humid, large food reserves, 
consisting of plant shoots and motley-grass— ephemeral grass formations 
are built up. This makes it possible for the animals to move from the 
310 Kabristan ridges through drying rivulet valleys almost to the sea coast, 
30-40 km beyond the outer limits of their fixed habitat. 

Meanwhile the first hot and dry season again causes mass mortality 
among the animals and a new withdrawal northwestward from their 
distribution area. This phenomenon is very significant in explaining the 
mode and rapidity of migration and the history of the animal range areas 
resulting from secular climate and landscape variations. 

In Talysh the vole may be found at altitudes of 1,500-1,800 m in the 
grassy mountain steppe, while it is absent in the poor fodder grouping of 
thorny astragali. A dense vole population developed on the eastern and 
northern slopes of Talysh near the upper timberline, which has been 
considerably lowered because of grazing and cutting. The upper parts of the 
Vilyash-Chai gorge in the Yardymly region and the area of Vassaru-Chai 
near Lerik which is planted to grain are the real foci of mass vole 
reproduction. Such foci are the results of human activity. From them, 
the vole penetrates downwards to the gorge bottoms of the eastern slopes 
of Talysh only when a forest has been thinned or completely cleared. 

In the middle Araks valley the vole is found in Aralik and on some of 
the tributaries of the Araks, such as Alindzha-Chai, Nakhichevan-Chai 
and Arpa-Chai. The vole is not observed on the sun-scorched slopes and 
ridges of the Zangezur Range in the zone of stony-mountain— saltwort-semi- 
desert and tragacanth astragali because of the poor nutritive value of the 
vegetation. In the central Armenian Highland the vole penetrated into the 
Araks valley by the left river bank. A steppe vole habitat is located in the 
upper part of the Pambak River valley, in the grassy mountain-steppe zone, 

* Ognev's report ( 1950) on the habitat of this vole in the Lenkoran lowlands is based on a confusion in the 
location of geographic points: the Kelvyaz outpost is in the high Talysh Mountain and not in the Talysh 
lowlands. 



306 



at an altitude of 1,200-1,800 m. This site is cut off from the west by the cold 
Akhalkalakhi Plateau, and fronn the north by the afforested spurs of the 
Bzovdal'skii ridge and the Pambak River gorge. Contrary to Ognev's 
opinion (1950), the range of this species bears no relationship to that of the 
Gori subspecies. Such a relationship should be sought to the south in the 
Leninakan area and Kars Upland where the dry mountain valleys probably 
have populations morphologically similar to M. socialis schidlovskii. 
Dal' and Zakharin (1951) indicate that this vole's distribution extends from 
the upper reaches of the Pambak through the Araks valley up to the latitude 
of Yerevan. 

The population of the steppe vole is greatest during the reproducing 
season, when the number of burrow openings observed per hectare attains 
9 0,000 on both plowed and unplowed land in the piedmont andropogon-steppe 
zones of the Adzhinour and Kartalinia plateaus, the eastern foothills of 
Karabakh and the northwestern slopes of Talysh. In this steppe landscape 
the contemporary ecological optimum of the species has been reached 
(Satunin, 1912a; Vereshchagin, 1942c, 1946b). 

Viable even in periods of strong inhibition of population, the colonies of 
this rodent are clearly associated with shrubs of Christ's-thorn which 
affords them a safe shelter from predatory birds and extreme heat. The 
juniper— pistachio forests adjacent to the beard-grass steppe —the so-called 
"arid stunted forest" — is sparsely populated by voles; they are encountered 
in thinned forest glades of motley- grass where they construct burrows under 
ephedra shrubs. 

The characteristics and the periods of steppe vole reproduction are 
dependent upon altitude and man-made landscape alterations. In the 
311 semidesert zone of the eastern Transcaucasian lowland, its period of 

reproduction is confined mostly to October-November and to March-April; 
although if the winter is warm and ephemeral vegetation available, it may 
extend from October to May. But in the mountain steppe of Talysh and on the 
Armenian Highland, intensive reproduction is observed in June, because 
of the later burning-off of the vegetation. 

The vole maintains longer sexual activity and vitality on unirrigated 
cereal croplands in the piedmont (Figure 124) and on irrigated cereal and 
alfalfa croplands in the semidesert zone of eastern Transcaucasia, than 
on virgin soil. 

Climate is the main factor affecting population variations and area of vole 
ranges, especially in eastern Transcaucasia. Nevertheless, epizootic 
factors, predatory mammals and birds also figure significantly in reducing 
the population of this rodent during peak breeding years. The greatest effect 
of bird predation on the vole population can be seen on the semidesert and 
the piedmont steppe from November to April, i.e. , during the migration 
and wintering of the birds (Vereshchagin, 1946b). 

Agriculture contributed to the expansion of the range of this species, 
opening new territories by the felling of lowland forests, and providing 
for an increased population by supplying a succulent and high calorie diet of 
cereal crops . 

The main morphological features of steppe vole (permeable fur 
covering, short extremities poorly adapted to rapid and remote migrations, 
flat skull, elongated diastema indicating its enormous burrow -digging 
capacity in heavy and medium soils) testify that the species was formed 
under topographical xerophytic conditions of the eastern Mediterranean. 



307 




FIGURE 124. М. socialis in a barley field 

Photograph by author, 1939 

312 The steppe vole of the Caucasus and eastern Asia never emerges in the 
desert proper but is dispersed in its eroded interzonal fringe areas, 
indicating an earlier association of presently separated steppes and semi- 
desert areas, probably during certain phases of the Pleistocene. The 
reasons for this association have been discussed elsewhere by us 
(Vereshchagin, 1942c; 1946b) and by Geptner (1945). 

Pidoplichko's opinion (1951) on the recent development of the vole range 
in the eastern Mediterranean rests on a patently erroneous description of 
the distribution area. 

The initial evolution of a variety of southern gray voles probably took 
place in the eastern Mediterranean on the relatively dry Pliocene uplands 
of Asia Minor. One of these forms, the Russian steppe vole, populated the 
dry coasts of eastern Mediterranean basins in the Lower Pliocene. 

The history of the development of the vole's range on the Caucasus 
follows orogenetic movements, fluctuations in the Caspian Sea level and 
shifts in landscape zones — both horizontal and vertical — and reveals the 
lability of the species in adapting to its environment. For instance, isolation 
of the Gori subspecies was caused by the rise of the northern spurs of the 
Trialet and the southern spurs of the Kakhetian ridges which partitioned off 
the Kura valley near the Mtskheti. M. socialis schidlovskii may 
be considered a special species having an earlier isolation. This isolation 
was accomplished by the uplift of the Armenian Highland in the Quaternary 
which spared some xerophytic communities in separated areas. Perhaps 
vast lava flows overlapping the sections of the ancient plateau in different 
directions contributed to this isolation (Vereshchagin, 1942c; Shidlovskii, 
1945). 

The range of the steppe vole on the Isthmus evidently fluctuated repeatedly 
during the Pleistocene, conformingto the development and reduction of 
landscapes suitable for habitation during different climatic phases. In our 



308 



time, forest clearing and slope erosion have facilitated the penetration 
of the species into the foothills and mountains, thereby greatly expanding 
its range. This is especially conspicuous on the mountain slopes of 
Karabakh and Talysh. The vole penetrates the semidesert and desert of 
eastern Ciscaucasia when channels are dug and new regions are developed 
for grain growing. New extermination methods have no noticeable 
reversible effects upon these incursions. 

Common vole — Microtus arvalis Pall. The fossil remains of this 
animal are known from many Pleistocene cave burials and from alluvial 
deposits in Eastern Europe, from Ireland to the Carpathians and from Italy 
to Sweden. In the U.S.S.R. the remains of this vole are common in Upper 
Paleolithic strata of the Crimea (Vinogradov, 1937b) and in Middle and 
Upper Pleistocene strata of the lower Don and the Urals (Vereshchagin and 
Gromov, 1952). 

On the Caucasus the bones and skulls of this vole were found in Middle 
Pleistocene strata of the Apsheron Peninsula together with those of the 
steppe vole (Argiropulo, 1941b; Gromov, 1952). ■•'- 

In the Recent epoch this vole inhabited the taiga, the forest steppe and 
even desert areas from the Atlantic Ocean to the Khingan and the upper 
reaches of the Amur. It has a peculiar distribution on the Caucasus 
(Map 71). 
313 In Ciscaucasia an almost unbroken distribution range runs from the Don 
steppes through Manych along the Azov shore and includes the lower reaches 
of the Kuban and the Taman Peninsula. 

This vole is widespread in the Stavropol uplands where it lives in 
mesophytic meadows and deforested areas. 

The northeastern limit of the M. arvalis range in the Ciscaucasian 
steppes approaches the longitude of Budennovsk, coinciding with the margin 
of semideserts in the lower Kuma area. Eastward it pushes far through 
the river valley to the Terek-Kumsa sands. Following the courses of the 
Terek and Sulak rivers as far as the Caspian coast, habitats of this vole 
occupy ridges and depressions with meadow, tugai and steppe vegetation. 
On the Ciscaucasian dry steppe, freshwater basins are a requirement for 
the maintenance of vole habitation (Naumov, 1948). On the Greater 
Caucasus this vole inhabits only the Trans -Kuban sloping plain in the 
foothills at an altitude of 900-950 m, and is not encountered in higher forests. 

In the mountain meadows of the Greater Caucasus, it is always replaced 
by the pine vole. 

The vole is common on the sloping plain of central Ciscaucasia, but is 
rarely found, if at all, in the mountains (Ognev, 1950). On the deforested 
sloping plain of Terek-Sunzha voles are numerous and apparently 
destructive. Turov (1926c) considered the species to be common only in the 
forest zone of North Ossetia. According to our observations, the deforested 
glades and subalpine meadows in the Fiagdon and Gizel'don gorges of the 
Lesistyi and Skalistyi ridges are heavily populated by this animal. It is 
common in the Urukh ravine at altitudes of 1,500-1,800 m, and, farther 
east, has been traced in the Argun and Sulak basins at altitudes from 800 m 
to 2,700 m (Berne, 1933; Geptner and Formozov, 1941). 

It has not been observed in the Dagestan foothills from Khasavyurt to 
Derbent, or farther southward on the low littoral terrace down to Baku. 

It is also absent from the humid regions of the Khachmas lowland which 
is inhabited by steppe voles, field and house mice. On the Kusary sloping 

• Lower jaws of this vole species were found in the Lower Pleistocene strata of Kudaro I in 1958. 

309 



plain, the lower border of the range is at an altitude of 600-650 m 
in the pasture-forest zone, and only on meadow sections of the Samur 
valley does it descend to 250-300 m. 

If accurate, Berne's mention (1928, p. 143) of animals caught near the 
Samur estuary is noteworthy. 

At the eastern end of the Greater Caucasus this rodent is encountered 
in grain fields and on subalpine meadows in the Konakhkent and Khizy 
regions. Its lower range in the Shemakha area extends along an altitude 
of approximately 800 m, while farther west on subalpine meadows in the 
upper reaches of the Gerdyman-Chai the species is replaced by the 
pine vole . 

In the deforested areas on the southern slope of the Main Range, the vole 
penetrated from mountain meadows through glades overgrown with bracken 
by way of the Demir- Aparan-Chai, Bum-Chai and Mazym-Chai ravines 
(in the vicinity of Kutkashen, Vartashen and Lagodekhi) up to the base of 
steep slopes and into the zone of lowland forests of the Alazan-Avtaran 
valley. On the high mountain meadows of the eastern half of the southern 
slope of the Greater Caucasus, the species is usually concentrated alongthe 
margins of cattle stands heavily covered with manure and overgrown with horse 
sorrel and orchard grass . It also populates subalpine meadows on the deforested 
ridges of the southern lateral ranges. 
314 The vole is common on meadows of the Kakhetian Range and in intensely 
deforested South Ossetia down to the reaches and tributaries of the Rion; 
here it is replaced by the pine vole as on the northern slope. 

There is no indication of the species in the dry Gori depression or in 
humid Colchis, on the Black Sea coast or on the northern slopes of the 
Adzhar -Imeretian ridges (Satunin, 1913; Shidlovskii, 1947, 1948, 1950). 

The common vole is conspicuous on the Trialet ridge and its spurs, 
e.g., the Mokrye Gory, and on Dzhavakhetia meadows. 

The species is eurychoric on the Armenian Highland, where along its 
lower limits the distribution increases greatly on the south and the east. 
While this vole is very common on ridges surrounding Lake Sevan (Turov 
and Turova-Morozova, 1928), it is also found on relatively dry and 
deforested northern slopes at an altitude lower than Kedabek (1,400 m). 
And similarly on the Karabakh mountain meadows (e.g., in Dali-Dag, Sary- 
Yeri and Basargechar) it is widespread and yet descends to 1,200-1,300 m 
on deforested eastern slopes. 

On the southern slope of the Zangezur ridge in the Akulis-Chai Basin, 
we found it in spring- fed marshland among heavily grazed steppe like spurs 
and valleys only above 1,800 m. The common vole does not occur in the 
tragacanth zone, i.e. , in the "phrygana" and "gariga" formation or in the 
artennisia and saltwort semidesert of the middle Araks valley and the 
Kura- Araks lowland. Contrary to Ognev's opinion (1950, p. 209) and 
Kuznetsov's maps (Bobrinskii, Kuznetsov and Kuzyakin, 1944), there is 
no trace of this vole in the Lenkoran lowland or on Mugan. In the high 
Talysh mountain it is encountered only at altitudes of 2,000-2,100 m on 
steppe meadows, i.e. , near the Kel'vyaz and Kosmalyan outposts. It does 
not even inhabit those pl£.ces inhabited by mole in the Talysh forest zone. 
In general, the present-day range of the common vole in the eastern half 
of the Caucasian Isthmus diverges somewhat from the distribution of the 
steppe vole. In northwestern Iran and northeastern Turkey, the common 



310 



vole is dispersed at altitudes of 2,000-2,500 m, and farther south its range 
is probably higher. Blanford (1876) recorded this vole under the name of 
Arvicola mystacinus De Fil. in an area between Shiraz and Esfahan, 
at an altitude oi approximately 2,800 m. =■= Goodwin recorded it from the 
eastern Elburz (1939). It probably inhabits Mt. Savalan and was caught 
in the region of Lake Urmia by the Caucasian Museum Expedition. In the 
Georgian Museum collections there are specimens from Kars, Oltu, Great 
Ararat and other localities in Turkey. The numerous vole populations 
in the high Eleskirt valley in the upper reaches of the Euphrates can be 
accounted for by the many swamps and rivulets. In central Turkey some 
remaining habitats are considered as relicts (Neuhauser, 1936; Ellermann, 
1948). 

With higher seasonal and daily temperatures, the factors of soil humidity 
and mesophytic vegetation assume greater importance for the rodent in the 
south, and result in an enlarged vertical distribution as the range extends 
southward. They also account for the division of the range into isolated 
relict islets adapted to high plateaus and ranges. The animal's range 
315 is obviously wedged out on the high mountain ranges of the nniddle courses 
of the Tigris and Euphrates. 

The absence of the animal from western Caucasia and Transcaucasia 
is adequately explained by the excessive moisture, while in the forest of 
the Talysh lowland it can be attributed to dryness of the soil and vegetation. 

The feeding habits of the common vole show it to be a typical herbivorous 
animal. Human intervention during historical time is responsible for the 
concentrated forage which comprises its present-day diet. 

The main natural factor limiting this vole's population in both mountains 
and plains is unfavorable climate. The influence of quadrupeds and 
predatory birds on this vole during the summer season on the Caucasus 
is considerably greater than on the steppe vole. 

Based on its contemporary range and ecology, it is assumed that the 
common vole is a European mesophilous species which developed under 
conditions of moderate climate and humidity on meadow vegetation. 

The explanation of the appearance and total isolation of large, common 
vole populations on the ranges of southwest Asia and the Caucasus is 
consonant with this concept if we presuppose an early settlement during 
a humid epoch when present-day hot desert valleys and slopes were well 
covered with meadow or, at the least, steppe vegetation. 

From an interpretation of the contemporary ecological and distributive 
characteristics of this vole, it would seem that it could hardly have appeared 
on the Caucasus before the Middle Pleistocene, and then during a time of 
considerable climate cooling and development of mesophytic meadow 
formations in the desert and semidesert zones. 

The origin of the local, high-mountain common vole in the mountains 
of the eastern Mediterranean and its subsequent settlement in the north 
is substantiated by the Mediterranean origin of northern mesophytic meadow 
formations. Some botanists now hold the theory of an alpine origin of 
meadows which were displaced to lower altitudes in the Pleistocene and 
advanced in a northerly direction (Fedorov, 1952). 

Pine vole — M icr ot us (Pitymys) majori Thos (s. lato). Voles of 
the subgenus Pitymys are known from Pleistocene deposits of Eurasia 

* A vole closely related to M. guentheri Danf. et Alst. is found on Mt. Talysh and in Iran. 



31i 



and North America (Wolf, 1939; Simpson, 1945). On the Caucasus remains 
of a special species of vole M. apscheronicus Arg. , probably ancestral 
to extant Caucasian species, are found in Middle Pleistocene strata of the 
Apsheron Peninsula. 

The distribution of pine vole on the Caucasus includes the Greater 
Caucasus, the Lesser Caucasus, Asia Minor, Talysh and Elburz (Map 70). 

In Ciscaucasia the pine vole can be found from the northern piedmont 
forest belt up to the talus passes on the Main Range. 

On the Trans -Kuban Plain the animal inhabits some locations to an 
altitude of 250-300 m above sea level, and it is common from Krymskto 
Maikop on the first low ridges of relict oak forests and mesophytic meadows. 
The range does not reach to the broader section of the Kuban River even 
in the valley of its left tributaries — here it is replaced by the common vole. 
On the east, the range enlarges and passes south to Cherkessk at an altitude 
of approximately 600 m. In the forest zone and on the alpine meadows of the 
northwestern Caucasus, the species is very common, and numerous 
specimens have been collected from the Caucasian and Teberdinskii 
reservations. 
316 In central Ciscaucasia the animal inhabits the mesophytic ravine areas, 
and is absent from the greater part of the dry Kabarda sloping plain. 

One isolated section of the range lies in oak stands in the Pyatigor'e 
area (Beshtau, Zheleznaya, Razvalka), and another more to the north 
in the Stavropol uplands. In this last area the vole inhabits grazed-down 
meadow glades among relict broadleaf forests, e.g., in the vicinity of 
Mt. Strizhament. 

The breach between this section of the range and the major one in the 
Caucasian uplands is 50-60 km and is made up of steppelike foothills. There 
are relict patches of pine vole habitation in forested dry ravines along the 
Kalaus. On the sloping plains of the Terek and Sunzha the animal inhabits 
mesophytic meadows which appeared shortly after the forests were cleared. 

Contrary to Ognev's opinion (1950), this vole is found neither in the 
vicinity of Mozdok nor near Grozny, where the climate is too dry for it. 
In central Dagestan it inhabits meadows at 1,600-2,500 m above sea level. 

In western Transcaucasia the pine vole is common everywhere from 
the Black Sea coast to the mountain talus passes. In humid Colchis it is 
absent only from periodically flooded alder thickets and reedgrass bogs. 
The Svanetian alpine meadows are heavily populated with vole along the 
fringes of birch forest grown with whortleberry and as far as the upper 
sections of cereal grass— crowfoot carpets which wedge into slate talus. 
Here it completely replaces the common vole. To the east, the animal is 
encountered all along the southern slope of the Greater Caucasus near the 
upper reaches of the Pirsagat rivulet in Azerbaidzhan. 

The pine vole is sometimes replaced by the common vole in alpine 
meadows in the areas of Nukha and Kutkashen. Argiropulo's statement 
(1939a) that the pine vole does not inhabit the forest zone in the Greater 
Caucasus is in error; on the contrary, a large pine vole population is often 
observed on the forest edges and in thinned stands of oak— hornbean. 

There is no indication of pine vole in the Alazan-Agrichai valley, in the 
valleys of the Adzhinour Plateau or on the Kura Lowland. Claims that this 
animal was found in Geok-Tepe [Geok-Chai] on desert foothills to the north 
of Evlakh (Shidlovskii, 1938; Ognev, 1950) are erroneous (Vereshchagin, 
1949c). 



312 



In eastern Taurus and along the ranges of the Lesser Caucasus the animal 
is spread mostly on the northern slopes and on the high humid plateaus of 
Dzhavakhetia and Tsalka. 

It is very common on deforested slopes of ravines of the Shakhdag Range 
and in the remaining beech forests of the Terter-Chai ravines in Karabakh 
where it descended to an altitude of 1,000-1,200 m. In Talysh the pine vole 
inhabits only the border areas of the upper timberline, and is rare because 
of the extreme aridity in summer (Vereshchagin, 1949c). The southeastern- 
most section of the range is the northern slope of the Elburz Range 
(Ellermann, 1948). 

Pine vole is very common on motley-grass meadows near the forest's 
upper edge. In these areas of ecological optimum, the greatest population 
of this species can be found. 

The pine vole resembles the common vole in feeding habits, but, because 
of its greater digging capacity, the rhizomic constituents of herbaceous 
plants feature more prominently in its diet. 
317 The population dynamics of this vole are considerably more stable than 
those of the steppe and common vole. Predatory birds inflict less population 
loss on the pine vole because of its subterranean way of life and its habitat 
in tall grasses and forests. 

Based upon the degree of fur -cover differentiation (underfur and guard 
hair), the pine vole is intermediary between the steppe and the common vole. 

The pine vole is a typical mesophilous species of the eastern 
Mediterranean formed on mountain meadows and in broadleaf forests. This 
species is an excellent indicator species which illustrates the long existence 
of mesophytic conditions on separate areas of the Caucasus-Asia Minor 
land mass. The peak in pine vole population coincides with the epoch of 
greatest cooling during the Pleistocene. 

The Stavropol-Caucasus break in range probably occurred at the 
beginning of the Holocene; the Lesser Caucasus -Asterabad break 
undoubtedly happened considerably earlier. 

Саиса81ал snow vole — Microtus (Chionomys) gud Sat. (s. lato) 
(Figure 125). Fossils of this vole were found during the excavation of the 
Kudaro caves in the reaches of the Rion. The animal is distributed within 
limits of 1,600 to 2,000 m on the Perevalnyi and Skalistyi ridges but is 
absent from the frontal chain of the Chernye Gory Mountains (Nasimovich, 
1935). 

Our observations in the Kuban and Teberda ravines show the vole at 
1,500-1,700 m above sea level on the Skalistyi slopes and especially 
numerous higher in the alpine zone on large rock taluses overgrown with 
pine, birch and willow, and further on in alpine meadows near taluses and 
glaciers. 

At present this species is not encountered on the Stavropol uplands 
(Vereshchagin and Gromov, 1953b), but it is found in the trachyte rock 
taluses on the permafrost section of the Razvalka Mountain in the Pyatigor'e 
area, at an altitude of 600 m. This population is now isolated from the 
main range 50-60 km to the north, since the nearest vole habitat is 
encountered in the canyon of the Berezovka rivulet, south of Kislovodsk 
(Vereshchagin, 1953a). Higher in the mountains, M. (C h i о n о m у s) gud 
Sat. inhabits the ravines of Malka, Baksan, Chegem and Cherek, especially 
rock taluses in canyons cut by these rivers in the Skalistyi ridge. It has 



313 



been observed in certain places in the interior longitudinal valleys at an 
altitude of 1,200-1,500 m, where there are stone taluses. This vole is rare 
in the Chernye Gory Mountains near Nalchik. 

In the Ardon ravine the heaviest population was observed on the lateral 
moraine in the pine— birch forest near the end of the Tseya glacier among 
granite rocks overgrown with fern, whortleberry, sweetberry, honeysuckle 
and rhododendron. 

In the Terek gorges, this vole is widespread on the slope of Mount 
Fetkhus near Ordzhonikidze (Formozov, 1926). 

In central Dagestan numerous voles are encountered in Gunib, in the 
Bogos Range and in the ravines of the Samur tributaries at altitudes of 
1,300, 1,600, 2,100 and 2,500 m (Geptner and Formozov, 1941). 

In August 1952, the author observed many of these voles in arid, hot 
depressions in the upper reaches of the Manas brook among marl taluses 
near the village of Lavashi at an altitude of approximately 1,000 m. Summer 
318 conditions here are pessimum for this animal. 



% jf-e 




FIGURE 125. Microtus (C h i о n om у s) g ud Sat. (s. lato) 

From watercolor by K.K. Flerov 

On the southern steep slope of the Greater Caucasus in eastern Trans- 
caucasia this vole is found in gorges at 1,300-1,400 m, e. g. , in the 
Gerdyman-Chai ravine and others, but it is common only in the cirques at 
river sources bordering birch— hornbeam formations at 2,000-2,600 m. It is 
also encountered on the ridges of the transverse ranges near the upper 
forest belt of mountain oak at 2,200-2,700 m above sea level. Farther west 
the vole appears on the Kakhetian ridge above Telavi near the snowline 
(Shidlovskii, 1951). In relatively dry and deforested South Ossetia. it occurs 
among the rocks from 2,000 m upward (Ognev, 1950). 



314 



In upper Svanetia this animal is very nunierous in fir— maple forests. 
In the arid Ingur valley it lives in sunheated gneiss rocks overgrown with 
Pontic azaleas, e.g., near the village of Mestia. On steep southwestern 
slopes in Abkhazia it is common from an altitude of 500 m upward to the 
glaciers, although Shidlovskii (1950) recorded it for high hiountains only in 
the reaches of the Bzyb River. In all probability, the Surami Range is not 
actually a habitat for this vole. In Asia Minor it probably inhabits the 
greater part of eastern Taurus since it was found on Mount Varzambeg 
south of Rize at an altitude of 3,000 m.(Neuha.user, 1936). It was recently 
observed by Shidlovskii in Dzhavakhetia as well. 

The southern range of this vole is not as wide as that on the Greater 
Caucasus. 

Taluses of big rocks held by birch and pine forests with undergrowth of 
whortleberry, honeysuckle, raspberry and Pontic azalea are densely 
populated by M. (Chionomys). 
319 The structure and placement of rocks and stones which ensure dry 

shelters and food storage are secondary in importance for the well-being 
of this animal only to the availability of the food itself. The composition 
of the bedrock is less important, inasmuch as calcareous, gneiss, granite 
and slate taluses are all found populated. 




FIGURE 126. Reserve stocks of hay stored by M icrot us (Chionomys) 

Photograph by author, 1952 

The composition of summer and winter foods of this vole varies greatly 
according to the habitat. Nasimovich (1935) observed that in the Caucasian 
Reservation the vole feeds mainly on anemones, ferns, raspberry and 
Caucasian rhododendron. In the upper part of the Teberda Reservation, 
Stepanov found that 32 vole stacks contained 47 plant species, of which 
ferns, raspberry, mountain ash, willow, birch, cereals, strawberry, 
spruce, pine and foxberry were the most frequently encountered. 



315 



According to observations made by Semenov-Tyan-Shanskii and the 
author in the Tseya ravine, the animals stored mainly twigs of honeysuckle, 
whortleberry, willow herb, mountain ash, Caucasian rhododendron and 
stone bramble. They also feed on branches of birch and shrubs, climbing 
very high on the smooth tree trunks. On the relatively dry meadows of the 
Doniserdon and Baksan valleys, stacks of hay, stored in fissures in fortress 
ruins (Figure 126), are composed of veronica, sage, nettle, clover, etc. 

The Caucasian vole is adapted to rock reliefs and avoids digging. Among 
the other voles, it stands out in its astonishing climbing ability, almost 
the equal of that of the dormouse or the squirrel. 

Notwithstanding the fact that great variations in the population of this species 
have not been noted, in certain years and in certain gorges they are rarely 
320 encountered. This may indicate the incidence of local epizootic diseases. 
This species is not as vulnerable to climatic affects as are other vole 
species. 

Of the predators, only marten, weasel and large owls are dangerous 
to this vole, although any newly-settled carnivore, like the sable, may prey 
upon it. 

Judging by the large population it supports and the size of the specimens 
it produces, the subalpine zone of the western part of the Greater Caucasus 
provides the optimal contemporary ecology for M. (Chionomys) gud. 

The formation of rocks during orogeny and glaciation was often more 
important to the development of this vole's range than the presence of 
mesophytic vegetation, as can be seen by the densely populated, dry 
interior valleys and by central Dagestan where the ecology of this species 
shows features similar to those ofMicrotus (Chionomys) 
nivalis Mart. (s. lato). An analysis of the range and habitat brings us 
to the conclusion that the Caucasian vole is an early local species, 
associated in its evolution with the development of high-mountain plant 
formations of the Miocene. The glaciation of the Greater Caucasus in the 
Pleistocene could not have displaced the populations of these animals to the 
plain, as thought by some faunists and paleontologists; it only partially 
lowered their range. Even during the lowering of the snowline to 1,000- 
1,100 m, a wide habitat for vole existed on the meridian ranges and on the 
fronta] longitudinal ranges. However, the cold and humid climate during 
the glaciation, which brought about the development of meadow formations 
in the foothill plains, created the conditions for the penetration of the 
animals into places below river bank escarpments. The vole probably 
appeared in the Pyatigor'e area during one of the cold epochs, but it is 
doubtful that it reached the Stavropol uplands. Future fossil finds in the 
foremontane regions from the Middle Quaternary may fix this history more 
precisely. 

Nowadays, cattle grazing and the felling of mountain forests diminish 
the range of the Caucasian vole in the eastern Caucasus, but enlarge it in 
the western Caucasus. 

Asia Minor snow vole — Microtus (Chionomys) nivalis Mart . 
(s. lato). Remains of this species were recorded from Pleistocene and 
Holocene strata of many caves in England, France, Germany, Italy, Austria, 
Sweden, Czechoslovakia and Hungary (Wolf, 1939). 

The remains of a special species — M. (Chionomys) machintoni 
Bate —found in Acheulean to Mesolithic deposits (Bate, 1937) are recorded 



1704 316 



for the caves of Mount Carmel in Israel. In the U.S.S.R. only remains of 
Holocene Age are known on the Armenian Highland. The contemporary 
species is widespread in Mediterranean mountains — from the Pyrenees 
to the Balkans, in Syria, Palestine and Asia Minor, on the Caucasus and on 
the western Kopet Dag. In the upper Pleistocene this vole probably inhabited 
even the Balkan Mountains. 

On the Greater Caucasus, the animal is known only from the northern 
slope of the range: the Fisht, Pshekish, Bolshoi Tkhach, Acheshbok and 
Pambak peaks, and from the upper Teberda and Cherek river gorges; it is 
not encountered in the Balkarian Cherek or farther east. The paleontological 
sites on the Greater Caucasus are located at altitudes of 1,500-2,000 m in 
the subalpine zone. The habitats here are the same as those of the Caucasian 
vole (Map 67). 
321 On the Lesser Caucasus the Asia Minor vole inhabits the forest and 

alpine zones on the Trialet ridge, in Dzhavakhetia, on all ridges of the 
Armenian Highland, in Karabakh and in Talysh. 

The vertical distribution increases in a southeasterly direction according 
to the locations of phytolandscape zones, although there are exceptions. 
On the Trialet ridge the species has been found at altitudes of 2,000 m near 
Bakuryan; on the Pambak ridge it is common at altitudes of 1,800-2,500 m; 
and on the Shakhdag it lives in the Shamkor-Chai ravine between 1,600 and 
2,300 m. Southward, it is encountered in the xerophytic mountain steppe, 
e.g., near Sevan on the Saraibulakh and Daralagez ridges (Turov and 
Turova-Morozova, 1928; Dal', 1944b, 1949b). Above Yerevan it occurs at 
a height of 1,200 m in rock taluses covered with shrubs (Flerov and Gureev, 
1934). 

Along the Akulis-Chai ravine on the southern slope of the Zangezur, in 
1947, we found the Asia Minor vole at altitudes from 1,600 to 2,500 m in 
rock taluses overgrown with spiraea, dog rose, almond, hawthorn and 
buckthorn in the zone of astragali-traganth. 

It was not encountered by the author in the beech forest in the Terter- 
Chai gorge in Karabakh, but is found higher in the alpine zone near 
Kelbadzhar . 

In Talysh and Kopet-Dag, there is only a relict range. The author found 
a small vole population in June 1945 on the bald Kelakhan Mountain in the 
dry Diabar depression at 1,900 m. The animals were encountered here in 
the taluses of teschenite rocks with a sparse xerophytic cover of saxifrage, 
oxeye daisy, brome shoots and xerophilic ferns among the mountain- 
saltwort semidesert (Vereshchagin, 1945b). This vole probably inhabits 
Savalan and the Elburz Range. 

The Asia Minor vole achieves its maximum population in a zone of steppe 
meadows, e.g., near the Sevan shore. The rocky substrate is required 
for the well-being of this species. The habit of storing food is not so 
strongly developed as it is in the Caucasian vole, but the capacity to adjust 
its habitats to varying conditions is immeasurably greater; it adjusts to a 
wide range of feeds and of humidity. Whereas in the subalpine zone of the 
western Caucasus this vole lives side by side with the mole and the earth- 
worm, in Talysh and on the southern slopes of the Armenian Highland it is 
found in the mountain xerophytic zone living with the tarantula and the 
scorpion. On the whole, the vole is more adapted to low pressures and 
rocky substrates and less to succulent foods and low temperatures. 



317 



The diminution in the animal's size in its eastern range and its 
disappearance from the medium -altitude southern slopes of the Armenian 
Highland indicate that the southeastern sections of the range are already 
located in a contemporary pessimum region. 

Discontinuities in the southeastern areas of this vole's range and the 
isolation of the mountain habitats in Talysh and Kopet Dag are explained 
by the increasing aridity in the eastern Mediterranean which followed 
optimal conditions for the settlement of the species during a pluvial cold 
epoch. 

However separated and various the isolated ranges in the Mediterranean, 
they indicate that the snow voles of the Pyrenees, Alps, Carpathians and 
Caucasus are separate species of parallel development from genetically 
related material. 
322 The postglacial rise in temperature and the elevation of phytolandscape 
zones have reduced the vole range, deepening it and isolating different 
geographic races of the snow vole. 

The limited spread of the snow vole on the Greater Caucasus is partially 
explained by the insufficient adaptation of this southern species to the 
marked humidity of the southwestern ridges; it is more difficult to explain 
its absence from the dry longitudinal valleys of the northern slopes of the 
eastern ridges and from Dagestan. 

Long-tailed snow vole — Microtus (Chionomys)roberti Thos . 
A unique finding of Quaternary fossils is recorded from the Upper Paleolithic 
of the Sakazhia cave in western Transcaucasia. The contemporary 
distribution of this vole includes the northern slopes of eastern Taurus in 
Asia Minor, the Adzhar-Imeretian Range, the Trialet ridge and the Greater 
Caucasus (Map 69). 

In the northwestern Caucasus this vole is found on the Chernye Gory 
Mountains from elevations of 800-1,000 m to the alpine zone. It is rare in 
central Ciscaucasian gorges and is not encountered at all on the Stavropol 
uplands. It is common in North Ossetia at altitudes of 800-1,000 m. 

In Dagestan it inhabits beech forests in the upper reaches of Andi Koisu 
near the villages of Takhota and Choroda, but it is probably absent from 
other parts of Dagestan (Geptner and Formozov, 1941), as it still is from 
the inner longitudinal valleys of the northern Caucasus. 

The author found the vole in Azerbaidzhan (Vereshchagin, 1940a) in 
1935 in the Katekh-Chai ravine, northwest of Zakataly, at an altitude of 
1,450 m, but it is not known farther east. 

The vertical distribution of M. (Chionomys) roberti Thos. on the 
BlackSea slope of the Greater Caucasus is from sea level to the upper 
forest belt (Shidlovskii, 1947, 1950). In Adzharia the animal inhabits the 
forest zone from 800 to 1,800 m (Shidlovskii, 1948). It is distributed in all 
of South Ossetia except the dry Gori depression. 

A typical well-populated habitat of this vole is a shady, humid, dead-end 
gorge. In the western part of the Greater Caucasus and on the Lesser 
Caucasus, such biotopes are to be found in fir and fir— beech forests with 
tall broadleaf grasses of burdock, cow -parsnip, nettle and orchard grass 
covering the bottoms of creek valleys. In Ciscaucasia the population is -more 
numerous in the lower humid area of the Chernye Gory Mountains than in 
the upper parts of large ravines. In eastern Transcaucasia the opposite is 
observed. 



318 



According to its morphological type, particularly its fur cover, this 
vole resembles the water vole to a certain degree. The animal often swims 
in mountain brooks and in general lives in conditions of extreme humidity. 

The speciation of this vole is associated with the early mesophytic and 
even hygrophytic phytolandscapes and biotopes which existed in the 
mountains bordering the Black Sea depression on the southeast and east 
in the Miocene. The reduction of the M. (Chionomys) roberti range 
to the east completely corresponds to the changes in landscape and biotopes 
in the mountain forest. 

The existence of an isolated, actually a relict, population in the upper 
323 reaches of the Andi Koisu in Dagestan can be explained by the penetration 
of the animal fronn the southern slope during a humid epoch. Tumadzhanov 
(1940) recorded a similar immigration of the beech into Dagestan in the 
Holocene. 

In all, M. (Chionomys) roberti is an endemic indicator species of 
the ancient mesophilous faunal complex in the Caucasus. 

Common red-backed vole — С le t hr i о nom у s glareolus Schreb. 
Fossils belonging to the genus are known in Europe from the Pleistocene. 
The main part of this species' range is located in taiga and broadleaf forests 
of Eurasia. The animal penetrates into the steppe through the forested 
areas of river valleys. Ecologically this vole is mesophilous and 
psychrophilic. Only the Pontic subspecies, С glareolus ponticus 
Thos. , is found in the southeastern Mediterranean, inhabiting the northern 
slopes of Taurus on the southern coast of the Black Sea (Thomas, 1906; 
Neuhauser, 1936) and the Adzhar-Imeretian Range (Map 72). 

In Adzharia the animal is found in spruce forests along the Supsa River 
(Shidlovskii, 1940a). This southern section of a wide range, now isolated 
by the Bosphorus, is a remarkable example of the southern faunistic 
influence on the Caucasus through the Balkans and Asia Minor. 

The common red-backed vole's absence from the western part of the 
Greater Caucasus indicates that it penetrated the southern Black Sea coast 
very late, during the period of maximum cooling in the Upper Pleistocene. 

Steppe lemming — Laguru s lagurus Pall. Fossils of this Upper 
Pleistocene species are known from several localities on the Russian 
Plain — the Don, Volga and Ural valleys. The present range is located in 
the steppe and semidesert zones from the lower reaches of the Dnieper 
to the upper reaches of the Yenisei and the borders of northwestern China 
(Vinogradov and Argiropulo, 1941). The steppe lemming is also widespread 
on areas adjacent to the Caucasian Isthmus, on the northern coastlands of 
the Black Sea, the Sea of Azov and the Caspian Sea. On the Isthmus it is found only 
in the northern Ciscaucasian steppes in the interfluve of the Kuma, Kalaus 
and Manych, i.e., northwest of the Stavropol Plateau (Sviridenko, 1928). 
Kistyakovskii (1935) found remains of this animal in bird pellets in the 
vicinity of Achikulak, Blagodatnoe, Sukhaya, Padina, Petrovskoe and 
Kambulat (Map 64). 

The steppe lemming appears in Ciscaucasia only in peak breeding periods 
or during migrations provoked by unfavorable conditions. We assume that the 
steppe lemming penetrated only recently into the Caucasus Isthmus, probably in 
the Upper Pleistocene or Holocene, from the adjacent steppes of the Russian 
Plain, drawing the inference from its underdeveloped range in the 
Ciscaucasian steppes, from its stable connections with the landscape, and 



319 



from formations of cereal grass— motley grass, feather grass— fescue and 
steppe— wormwood of the temperate belt of Eurasia. This settlement could 
have taken place at the time when the Manych strait disappeared. 

Promethean vole — Prometheomys schaposchnikovi Sat . 
(Figure 127). Fossils of this vole carried in by the eagle owl are known 
from Upper Paleolithic strata in the Gvardzhilas cave in the Kvirila River 
ravine in eastern Transcaucasia at an approximate altitude of 450-500 m, 
and from Lower Paleolithic strata of the Kudaro I cave. 

The Gvardzhilas cave is located outside the contemporary range of this 
animal, but to study the shifts in range during the Upper Pleistocene, it is 
necessary to follow the features of this vole's contemporary distribution 
on the Rachin ridge and to investigate the radius of the eagle owl's hunting 
224 flights in this region. 




FIGURE 127. Promethean vole 

Watercolor by E.Ya. Zakharov 



320 



325 



The present distribution of Promethean vole is confined to the western 
half of the Greater Caucasus and to the Adzhar-Imeretian Range of the 
Lesser Caucasus (Map 65). On the Greater Caucasus it is encountered 
from the Fisht and Oshten peaks to the Kazbek-Krestovyi Pass and Gudaur 
Dag. It has been traced in the Caucasian Reservation in the Mzymta valley, 
along the lower Kardyvach Lake, in the meadows where the Kurdzhips and 
Pshekish rivers rise, in the Kisha and Urushten basins, in the Malaya 
Laba sources, and near Krasnaya Polyana (Turov, 1926a, 1926b). The 
animal was caught in Abkhazia at the Avatkhar site at an approximate 
altitude of 1,500 m. In Svanetia in 1948 the author followed the animal on 
high-mountain meadows in the upper reaches of the Ingur and the Rion and 
found it particularly numerous on the Zagar pass of the Lechkhumi ridge 
at altitudes of 1800-2,800 m. This rodent is not encountered in the dry 
longitudinal Ingur valley. In South Ossetia the Promethean vole was hunted 
in the Bolshaya Liakhva upper reaches (Shidlovskii, 1951). The animal 
was traced on the mountain meadows of the Mamisson pass to the railroad 
stations Kazbek and Gudaur in the central part of the Greater Range (Turov, 
1926a, 1926b). The author found neither the animal nor its tracks in the 
upper reaches of the Balkar Cherek, Doniserdon and Ardon. We have no 
data on the presence of this vole in the upper reaches of the Kuban, Baksan 
and Malka. Thus, the range on the northern slope of the Main Range 
is broken at least from the Laba basin and Zelenchuk to Ardon. This may 
be explained to some extent by the destructivity of mountain glaciers. 

On the meadows of the Skalistyi ridge in Ossetia, in Kabarda and in the 
Grozny Region, this rodent was not observed in places perfectly adapted 
for its habitation. Apparently, the Promethean vole was not distributed 
north into the dry longitudinal valleys even during the Pleistocene. 

Nor has it been observed east of Kazbek on the Perevalnyi Range. In the 
Zakataly Reservation, which the author has carefully explored, it is absent, 
as it is from central Dagestan. The rapid thinning out of the rodent 
population east of the Caucasus can be attributed to the great reduction in 
area of alpine meadows and to xerophytization as shown by the development 
of sheep's fescue and matgrass, instead of a motley-grass community. 

On the Lesser Caucasus, the Promethean vole has been found only in 
the upper reaches of the Bakvis-Tskali River, the left tributary of the Supsa, 
near the Bakhmaro health resort, at an altitude of 1,600 m (Shidlovskii, 
1940a, 1947); these subalpine meadows are located west of the Gotimaria 
(Metis-Tskaro) peak (2,868 m). The range in this area is considerably 
wider; according to Shidlovskii 's data, the vertical distribution lies within 
limits of 1,500-2,800 m, but the animal can also be found at lower altitudes. 
The contemporary ecological optimum of this animal is within subalpine 
and alpine meadow zones with a predominance of lady's mantle and bunch 
grasses (orchard grass, meadow grass and others). Meadows lying among 
thickets of Caucasian rhododendron and birch are also densely populated. 
The rodent does not enter the rhododendron growth where food is not 
available, but in birch forests and on the upper forest edge grown with 
tall grass it is as common as it is on open meadows. This vole's habitation 
of beech forests in the Caucasian Reservation, noted by Turov (1926b), 
may be, in some cases, the consequence of recently advancing forest 
formations into the alpine meadow zone. In heavily shaded beech and 
hornbeam forests the vole is never found because of the lack of grass food. 



321 



In many places, the animal inhabits oat, wheat and potato fields or invades 
their borders, as in the upper reaches of the Ingur. 

Its strictly subterranean way of life and its diet of grass roots are 
reflected in the animal's morphology: flattened skull, extended incisor 
curve, elongated diastema, rudimentary eyes and shortened extremities. 
The skull, with a crest extending nearly to the occiput, is very similar to 
that of the muskrat. In its weakly-differentiated and vertically-disposed 
fur cover and coloration (often black), the animal is similar to the mole vole. 

In general, in a morphological-functional sense, the Promethean vole 
combines features of the muskrat, water vole and mole vole, but, in 
contrast to the latter, it lives in mesophytic conditions (see also Vinogradov, 
1926; Ognev, 1926b). According to Gambaryan's observations (1952), the 
Promethean vole, unlike the mole vole, combines terrestrial feeding with 
root grubbing. 

Such adaptations indicate the early origin of the species and its pristine 
associations with the landscape of the high-mountain mesophytic meadows. 

The population fluctuation of the Promethean vole can be attributed to 
mass death during the spring showers and after frosts when the water 
freezes in the burrows (Turov, 1926a). 

The strict endemism of this species shows the conservation of the 
ecological conditions on the western Caucasus: the ancient origin and 
continuous existence of the landforms and the zone of mesophytic subalpine 
and alpine meadows from the Lower Pliocene at the latest. 

The division of the range into two parts — the Caucasian and Lesser 
Caucasian — may have taken place by postglacial time as a result of the 
drying-out of the meadows on the Surami crystalline rock massif. 
326 This vole is an early Pliocene species endemic to the Lesser and 
Greater Caucasus with a relict range which has increased somewhat in 
present time because of the expansion of the high-mountain-meadow zone 
through forest-felling and animal-grazing. 

Water vole — A r vie ola terrestris L. (s. lato). Fossils of this 
vole are known from Pleistocene strata in Europe. 

Remains of A г vi с о la terrestris L. (s. lato) are common on the 
Russian Plain in Upper Pleistocene and Holocene deposits. South of the 
Caucasus the remains of this vole have been found in Acheulean strata of 
Syrian caves. On the Caucasian Isthmus there have been isolated fossil 
finds in Middle Quaternary strata of eastern Transcaucasia and in the Upper 
Paleolithic of western Transcaucasia, and numerous finds in Holocene 
deposits in caves which are day rests for eagle owls. 

The contemporary distribution of the species includes the forest, steppe 
and semidesert zones of Eurasia. South of the Caucasus it can be traced 
in northern Iran and Asia Minor as far as Palestine. 

On the Caucasus the water vole inhabits small areas from the level of 
the Caspian to an altitude of 2,600-2,800 m. but is not found in broadleaf 
mountain and coniferous forests, or in rocks and taluses of passes (Map 66). 
Its greatest population is observed on the floodplain meadows of the lower 
Don and a somewhat smaller population in the Kuban floodplain. 

In the Colchis and Lenkoran lowlands it is very rare, although it is found 
in damp depressions which flood annually. In the eastern Ciscaucasian 
semidesert and in Transcaucasia, it is common on lakes and streams 
overgrown with reed and rush, where it lives in the summer months and 
builds spherical nests amongthe thickets, but does not approach the shore area. 



322 



The largest population of water vole in the foothills is observed on spring 
rivers like those on the Terek-Sunzha Plain. 

In the interior steppe valleys of the Greater Caucasus northern slopes 
and in central Dagestan, a peculiar type of vole, ecologically distinguished 
by its subterranean existence, is found. In these areas this vole inhabits 
the bottoms of gorges, particularly irrigated farmland. In the alpine zone 
of the Greater Caucasus it occurs rarely and only in the upper reaches of 
the Terek and the Aragva. 

On the Armenian Highland, in Asia Minor and in Iran, the water vole 
is common on subalpine meadows on the upper reaches of rivers (e.g., 
in Karabakh, in the upper reaches of Akstafa and in the lake region of 
Dzhavakhetia). In the Araks valley many of these voles are found near the 
spring rivers of the Aralik depression. 

It can be seen that this vole is widely distributed and especially attracted 
to water biotopes. 

The ecological optimum of the subterranean populations of these voles 
is at mean altitudes in the dry area of Caucasian plateaus; that of the 
hydrophilous populations is in the semidesert zones which lack excessive 
atmospheric humidity [sici]. The low population density of the water vole 
on the Black Sea coast and in the Colchis and Asterabad swamps indicates 
a certain degree of similarity to the distribution of the common vole. 
In the Colchis swamps it is the tremendous competition and persecution 
from the Norway rat that produces this reduction in population, and in 
Asterabad the drying-up of reservoirs in summer and the compaction of 
the Lenkoran yellow podzol have the same deleterious effect (Vereshchagin, 
1941b, 1949d). 
327 Subterranean mountain populations differ from hydrophilic species in 
elongated diastema, greater curvature radius of the upper incisors, and 
shorter tail length, and the divergence in the contemporary ranges of the 
respective species can be observed along a line of secondary remove from 
the water medium and from these distinctions in features. 

In southwest Asia and on the Caucasus the formation of these characteristics 
is probably related to the uplift of mountain systems during the Quaternary. 

Paleoecological data suggest that the water vole of the Caucasus is probably 
of Upper Pliocene origin. In Transcaucasia the southern, relatively 
xerophilous populations actually predominate. 

The consistency of the species range in the Mediterranean is ensured 
by both the natural and artificial development of freshwater biotopes and 
the ecologic plasticity of the animal. 



Order PERISSODACT YLA 

Representatives of the families Equidae and Rhinocerotidae appear in the 
Quaternary fauna of the Caucasus, but by the Holocene only Equidae remained. 
The historical origin and disappearance of this family is examined below. 



Family EQUIDAE 

The earliest fossils of the large, true horse, Equus stenonis Cocchi, 
are known from the Caucasus and adjacent territories from Upper Pliocene 

323 



deposits. Probably the horse appeared simultaneously with the southern 
elephant introducing a new phase in the fauna. Later, in the Middle 
Pleistocene, the "semi-ass" appeared on the Caucasian Isthmus. 

The history of true horses can now be traced on the Caucasus very 
schematically from a number of deposits of various ages. 

Quaternary horse — E q u u s caballus (s. lato). After Equus 
stenonis, remains of which were found in Akchagyl and Apsheron deposits 
of the Isthmus, the Taman Upper and Lower Pliocene horse, E. aff. 
siissenbornensis Wiist . , found in conglomerates of the Taman 
Peninsula and the ancient Caspian terrace shingle near Baku, is the closest 
to the Quaternary horses (Chapter II and Map 79). 

Fossils of Midde Pleistocene horses (which are not recorded in detail) 
are abundant in Binagady asphalts and are found isolated in clayey beds of 
the Khazar stage in many localities in the Stavropol area and the Krasnodar 
Territory. These horses were related to E. caballus chosaricus 
Grom., found in Middle Pleistocene deposits of the Russian Plain. 

Remains of horses from the Il'skaya Paleolithic settlement in Ciscaucasia 
and from the Paleolithic sites of Imeretia and the Lesser Caucasus uplands 
are probably related to E. caballus latipes Grom. of the Russian 
Plain (see Gromova, 1949). South of the Caucasus, remains of Upper 
Pleistocene horses are known from Paleolithic strata in the Tamtama and 
Bisotun caves, from talus loams in the Maragheh vicinity of Iran and, 
328 southwest, from Palestinian and Syrian caves. The southern horses were 
of a lighter build than the northern animals. 

Specimens of Pleistocene horse have been taken in the following 
proportions to other species from sites representing various types of burial: 
6.6% from the Il'skaya encampment; 5.2% from the Binagady asphalt; 
2.5% from the Sakazhia cave; 16.5% from the Mgvimevi shelters; 50.0% 
from the Zurtaketi encampment. 

Assuming that the forms of horse we have mentioned, which inhabited the 
Isthmus during the Pleistocene, constitute a unique genetic series, then 
the origin of the Caucasian Holocene horse should not be sought in migrated 
forms originating elsewhere. 

It is difficult to separate the remains of wild horses which lived on the 
Caucasus in postglacial time from the remains of domesticated horses. 

In the strata of post-Paleolithic encampments, it is quite common to 
find bones of horse among kitchen middens and skulls and entire skeletons 
in the deposits. It is very probable that most of these remains belong to 
tarpans— E. caballus gmelini Ant . — or to their domesticated offspring. 

The records of horse remains found on the Caucasus in kitchen middens 
of encampments and in deposits from the historical age show that this animal 
had an enormous economic, dietary and ritualistic importance for various 
tribes and peoples (Table 7l). 

Numerous remains of horse are known also from Urartu burials of the 
10th to 12th centuries A. D. in the Sevan Basin (Lalayan, 1929) and from 
the cultural strata of the Later Bronze Age and the Middle Ages near 
Samtavro and near Mingechaur. * The horse (including the wild horse) 
was among the preferred subjects of Bronze Age drawings and sculpture 
in Southwest Asia and in the Caucasus. 

In the middle of the first millennium B.C., wild horses still inhabited 
the Caucasus in large numbers. These were thin-legged animals with little 

* Materials from these localities have not been preserved. 

324 



ears on small heads similar to the Arab horses. Their slight build and 
refined form had nothing in common with the body type of tarpans. 
Marvellous reliefs on the walls of the Assyrian palace in Nineveh (Atlas 
of Ancient History of Egypt, Southwest Asia, India and China, 1937, Table 
156, Figure 2) show hunts in which they were set upon by dogs and brought 
down by spears. This accumulation of evidence confirms the fact that the 
tarpanwas an aboriginal horse of Europe and the northernhalf of Asia and not 
a settler from the south at the end of the Pleistocene (much less during the 
Holocene) as is held by some paleontologists. 

In the monuments of the Koban cultures in Ossetia and Abkhazia the 
horse was depicted on belt buckles and ax handles as a rather slender and 
long-tailed animal. Bronze figurines of horses with heavy manes and long 
tails are among the material obtained by Uvarova (1900) in Ossetia. 
Without taking into account the stylization of the representations, one sees 
similarity in body frame, particularly in the light head of this animal, 
to the southern type of horse, in marked contradistinction to the tarpan. 



(329 



TABLE 71. Number of kitchen and ceremonial remains of horse, compared with other animals, in 
strata of encampments and burials on the Caucasus* 







Proportion of horse 


Regions and localities 


Dating 


specimens to 
other specimens 








in %) 


Ciscaucasia and Russian Plain 








Sarkel fortress (Slav and Khazar strata) 


8- 13th centuries A.D. 




10.3 


Settlement near Tsimlyanskaya (Khazar strata) . . . 


8- 10th centuries A.D. 




3.7 


Settelement of Zeyukovo near Nal'chik 


6- 8th centuries A.D. 




16.7 


Ancient town sites: Isti-Su and Alkhan-Kala in the 










2- 3rd centuries A.D. 




7.1 




6th century B.C. - 2nd 
century A.D. 








21.2 


Site of ancient town of Alkhaste on the Assa 


6- 5th centuries B.C. 




13.0 


Site of ancient town of Semibratnoe on the Kuban . , 


5th- 1st centuries B.C. 




17.0 


Cepi 
Ancient Greek towns of the 

Phanagona 
Taman Peninsula: ^ .,, т 
. Taman [HermonassaJ 


■5th- 1st centuries B.C. 




22.5 
12.3 
14.1 


Burial on the Manych 


1st millennium B.C. 




12.5 


f mud-hut 1 
Tsimlyanskaya on the Don: < j y, о 


Second half of 2nd 
millennium B.C. 




13.2 
12.0 


Eastern Transcaucasia 










12- 14th centuries A.D. 
9- 12th centuries A.D. 




16.6 


Settlement in the region of Baku fortress 


8.1 




1st millennium B.C. 




8.3 


Western Transcaucasia 




Akhshtyrskaya cave near Adler 


Neolithic and later strata 




2. J 



" Author's unpublished data. 



325 



In large Scythian mounds in Ciscaucasia and south of the Russian Plain, 
archaeologists often encountered mass ritual burials of horses composed 
of some tens and even hundreds of specimens (as many as 300). 

In Scythian memorials realistic representations of horses are quite 
common. A drawing of a stallion with a short-standing mane, thick legs, 
'^^^ heavy head, small ears and blunt muzzle resembling a tarpan is depicted 
on a silver dish from the Maikop burial (Farmakovskii, 1914; Figure 128). 
The well-known horse representations on the silver vase from the Kul-Oba 
burial probably portray domesticated horses. Outline drawings of horses 
are also found on the rocks of Kabristan, south of Baku. These were made 
probably during the time of the ancient Romans or earlier when wild horses 
lived on the Kura steppes (Vereshchagin and Burchak-Abramovich, 1948). 

Pallas, Gmelin and Guldenstaedt reported on tarpan in southern Russia 
and on the Ciscaucasian steppes at the end of the 18th century . Levchenko 
(1882), Keppen (1896), Anuchin (1896), Geptner (1934b), Antonius (1938) 
and others published many articles and notes tracing the history of the 
disappearance of this animal in the second half of the 19th century. A 
particularly complete and impartial survey on the investigation of the tarpan 
was submitted by Keppen, who insisted on the necessity for studying horse 
330 skeletons from the southern Russian burials. This investigation has not 
been carried out however. * 

Unlike most investigators, Anuchin did not consider the wild horse of 
the Eurasian steppes and deserts as a species, but thought it to be a type 
of horse which belonged to nomads and ran wild. It is obvious that, if this 
were the case, these horses ran wild on a large scale, and such horses 
must have closely resembled the original form — the wild horse of the Upper 
Paleolithic and the Lower Neolithic. There is little data on the tarpans 
in the Caucasus. Rovinskii (1809, p. 131) wrote: "Wild horses or tarpans 
are found in flocks in the Kuban steppe, near the Ural mouth and farther. " 

The author could find no historical data on the presence of wild horses 
in Transcaucasia. By the Middle Ages, the only survivor of the dense 
early population of wild horse on the northern plateaus of southwest Asia 
was the kulan, since the horses had probably been caught and domesticated 
by Hittites, Assyrians and Urartus long before the Christian era. 

The rapid evolution of horses in the Quaternary is usually associated 
with the subsequent development of steppe topography and coarse grasses, 
some of which — the grain formations — appeared as early as the Pliocene. 
The adaptive features and contemporary ecology of wild and 
331 domesticated horses corroborate this association. Notwithstanding this 

typical association, horses could still subsist on the forest steppe and even 
in the forest zone of Europe because, in large herds, they can actively 
defend themselves against packs of predators (e.g. , wolves), and because 
they can grab for food under the snow cover. In these areas, they fed in 
glades and swamps just as the extant domesticated horses of Yakutia do. 

The characterization of the Quaternary horse as a stenotopic steppe 
animal did not do justice to its conformation, which represents a universal 
form in ungulates. This is substantiated by the success in using domestic 
horses in widely differing topographies — from the glaciers of Greenland, 
Antarctica and the Alps to the tropical forests, swamps and deserts of all 
continents. 

* Skulls and skeletons of horses from earlier diggings in large burials in the Ukraine and Ciscaucasia have not 
been kept. 

326 



(330) 




FIGURE 128. Representation of a horse, primitive bulls, lions and other animals on a silver vessel 
from Maikop hill (Farmakovskii, 1914) 

Recently Vetulyani (1952) developed Keppen's theory that, especially 
in Middle Europe, wild horses were forest animals. As Vetulyani saw it, 
the tarpan adapted itself to a middle -forest-belt habitat and, in Europe, 
became a forest animal only during the last stage of its development, i.e. , 
in the historical epoch. 

New observations on artiodactyls would indicate that such a change in 
biotopes resulted from the development of forests in former forest-steppe 
areas and the hunting down of horses by man. 

The intersecting relief and relatively dense population of the Caucasus 
did not favor such adaptations. The steep slopes, both afforested and 
deforested, were unsuitable for horses, and the gentle slopes were too 
exposed. 

The date of the disappearance of the tarpan from the Russian Plain 
has been previously established as the 1870's, but Geptner's recently 
published data (l934b, p. 482; 1955) on the last tarpan to inhabit the Poltava 
Region considerably advances this date to 1918-1920. 

Tarpan extinction on the Caucasus resulted from their harassment by 
nomadic horsemen, steppe plowing, human settlement near watering places 
and, finally, from hunting with firearms (see Keppen, 1896). 

Kulan-Equus hemionus Pall. (s. lato). The fossil remains of 
Equus hemionus Pall, are known from Pleistocene and Holocene strata 
of Eurasia, particularly in steppe and semidesert zones. In Gromova's 
opinion (1949) kulan fossils recorded from Western Europe as 



327 



E. hemionus fossilis Nehr. may be related to small horses. However, 
the possibility of an early penetration of kulan into the west from Asia 
during the postglacial transformation of plains into steppes cannot be 
excluded. Phalanges and metapodia of semi-asses from upper layers of the 
Afontova Gora [Krasnoyarsk area] on the Yenisei, from sand dunes near 
Troitskosavsk [now Kyakhta] on the Selenga, from Irtysh and from Altai 
caves (Gromov, 1948; Vereshchagin, 1956) are mainly from the Upper 
Pleistocene and Holocene. Fossils from the Upper Pleistocene of China 
have also been recorded. 

In central Asia remains of kulan are known from Paleolithic deposits 
in the Bukhtarma cave (Vereshchagin, 1956) and from Bronze Age 
strata in the Kaunchi-Tepe cave [Tashkent area]. Duerst (1908) recorded 
the remains of a thin- legged horse, E. caballus pumpellii Duerst, 
from the earliest Annau strata (3000-1500 B.C.) near Ashkhabad. Gromova 
(1949) is of the opinion that these animals may be kulans . 

Kulan lived south of the Caucasus from the Lower Paleolithic Acheulean 
up to the present (Bate, 1937). Kulan fossils are also recorded from 
Paleolithic strata of the Barda-Balka cave in south Kurdistan. 

On the basis of contemporary data, the wild ass (Figure 129) is 
undoubtedly a southern species which originated in the Eurasian semidesert 
and upland steppe. 

The vague relationship of the Pleistocene ass, E. hidruntinus Reg., 
with the wild ass makes it difficult to establish when and by which routes 
the wild ass emerged on the Caucasian Isthmus. Up to now a direct 
phylogenetic link between these species was generally denied on the grounds 
that the wild ass had a more primitive dentition (Stehlin and Graziosi, 1935; 
Gromova, 1949). If we consider as an analogy related orders of Quaternary 
elephant, rhinoceros, horse and primitive bison, it is possible to conclude 
that the wild ass is a direct descendant of the Pleistocene ass. The claim 
of an autochthonous origin for the wild ass on the Caucasus follows from 
this conclusion. Remains of Holocene wild ass are known from several 
localities on the Caucasian Isthmus and in adjacent areas of the Russian 
Plain (Map 78). The oldest finds are from a settlement of the later half of 
the second millennium B.C. near Dzhemikent, south of Makhachkala; the 
latest finds are from 13th-century strata of a settlement in the vicinity 
of Baku bay. 

Historical data on the wild ass on the Caucasus and in adjacent territories 
are fairly numerous. Xenophon (1896 edition) gives a vivid description of 
ancient Greek wild ass hunts; these wild asses lived together with ostriches 
in the wormwood steppes of the upper reaches of the Euphrates. The records 
of Moisei Khorenskii (1893 edition) state that Armenian kings of the 
Arsacid Dynasty (lst-2nd century A. D. ) hunted onager (wild ass) and boar 
in the middle Araks valley. These records were recently discussed by 
Sarkisov (1946a) who mentioned that onagers had been recorded in Armenia 
in the "Code of Laws of the Armenian-Gregorian Church" in the Geography 
of Armenia by Ananiah of Shirak(7th century) and by Grigor Narekatsi 
(llth century). 

Under the influence of Satunin's ideas, Sarkisov assumed that the 
mountain barriers eliminated all penetration routes into Armenia for the 
onager except one from the "Aral-Caspian lowland. " Clearly, mountains 
are no barrier to this species, since, in historical time, onagers inhabited 



328 



every intermontane valley that was even slightly accessible, including the 
uplands of southwest Asia. 

Even more convincing are the arguments to substantiate the existence 
of the wild ass in Armenia in the book of mathematics by Ananiah of Shirak 
which is discussed by Ter-Pogosyan (1947). 

The Armenian historian of the 10th century Moisei Kalankatuiskii 
(1861 edition) mentioned the wild asses inhabiting the Barda kingdom 
(in the contemporary Mil'skaya-Karabakh steppe). Nizami Ganjawi 
(12 century) praises the wild ass hunt in the Kura-Araks lowland. Rashid 
ad-Din (14th century, 1946 edition) mentioned that among the victims of the 
grand hunt organized by Gazan Khan in Talysh were wild asses. At the 
be ginning of the 17th century Adam Olearius (1870 edition) recorded an 
onager hunt by the Iranian shah on onagers or Persian "gurs" in one of the 
sections of an enormous reservation in the vicinity of Iran's capital, 
Esfah^n. 

More recent data on the wild ass developed by Russian naturalists 
indicates that these animals inhabited only the area beyond the Volga during 
the time of Pallas . 

Levanevskii (1894) gives a good description of the habits of this species 
in the region of the Kirghiz steppes of the Emba District. From this 
description it seems that the wild ass was common in the semidesert 
between Emba and the Urals even in the second half of the last century. 
333 It was probably at this time that the wild ass population declined rapidly 
and the range shifted to the east and southeast. It is difficult to relate this 
shift and reduction in population only to the extermination of the animal 
with improved firearms or to an increase in human population. The Nogai 
and Kazakh retained their main weapons, percussion muskets and flintlocks, 
from earlier times. The increase of encampments was limited by the fodder 
capacity of desert and semidesert pastures. It is therefore probable that 
other causes influenced the dynamics of the aninnal population. 

By the beginning of this century the wild ass was found in the U. S. S. R. 
only in the semideserts of northeastern Kazakhstan — Bet-Pak-^Dala and 
Semirech'e. These animals disappeared from these areas and from 
southern Turkmenistan-Badchis in the 1930's. The peculiarities of wild 
ass distribution in the Iranian uplands and in Arabia are as yet unknown. 

Taking into account the stenotopic character of the wild ass and its 
adaptation to desert-steppe topography, one can fix its appearance on the 
Caucasian Isthmus at the beginning of the xerothermic period. It may have 
penetrated northwards into Transcaucasia from the valleys and low 
ridges of northwestern Iran and into Ciscaucasia directly from the Russian 
Plain and indirectly from central Asia by way of the northern coastlands 
of the Caspian. 

The broadest range of the wild ass existed in the Bronze Age. It is 
possible that the extinction of the wild ass in Transcaucasia coincides with 
the Mongolian invasion in the 13th century, and with the later growth of the 
Iranian [Persian] kingdom when the animals of the semidesert were 
destroyed in massive hunts. In northeastern Ciscaucasia wild ass survived 
until the 19th century. 



329 




FIGURE 129. Kulans in the Pleistocene semidesert of eastern Transcaucasia 



Order ARTIODACTYLA 

Representatives of the families Suidae, Camelidae, Cervidae and 
Bovidae inhabited the Caucasian Isthmus in the Quaternary. Pig, deer and 
334 Cavicornia existed there up to the Holocene. Paleontological and 

zoogeographical data will be examined in order to analyze the history of 
the range development and the evolution of Holocene species of artiodactyls. 



Family SUIDAE 

The fossil remains of Suidae, a bunodont nonruminant family, are known 
in Eurasia from the Lower Oligocene. The genus Sus, including the extinct 
subgenus Microstonyx, dates from the Lower Pliocene. Most of the 
fossil remains of Suidae were found in Upper Miocene and Pliocene deposits 
of the Mediterranean and southeastern Asia. 

Boar — S u s scrofa L. (s. lato). Remains of pigs belonging to the 
polymorphic species S. scrofa occur on the Caucasian Isthmus from 
the Upper Pliocene. The ancestor of the present-day boar was probably 
the large Taman boar, S. tamanensis (Vereshchagin, 1951d). 

The next stage of evolution is represented by remains of the Apsheron 
boar, S. apscheronicus Burtsch. et Dzhaf . , from Middle Pleistocene 
loams of eastern Transcaucasia, recorded by Burchak-Abramovich and 
Dzhafarov (1948). 

In the Akhshtyrskaya cave on the Black Sea coast, boar fossils were 
found in all strata from Mousterian to late Iron Age, their frequency 
increasing with time. Thus, in Mousterian strata they comprised 0.06%, 



330 



in Upper Paleolithic 1.2%, and in post -Paleolithic 20.0% of the total 
number of osseous remains in each strata. 

Boar existed in pre- and post-glacial time in Western Europe (Amon, 
1938). North of the Caucasus on the Russian Plain, there have been single 
findings of boar in the Upper Pliocene and Middle Pleistocene deposits of 
the middle Volga region. During the Pleistocene the boar was probably- 
very rare even in the river valleys of the Russian Plain, penetrating the 
area sporadically from the south in winters with a low snowfall. The rich 
fossil collections from the banks of the Don and the Volga contain only one jaw 
fragment of a Pleistocene boar (found on the Tunguz Peninsula), while from 
the Ural valley only remains of Holocene boars are found (Vereshchagin 
and Gromov, 1952). In general, a number of related facts indicate that 
the migration of boars from the Caucasus and their widespread settlement 
in the Russian Plain occurred only in the Holocene. 

South of the Caucasus, in northern and central Iran, Palestine and Syria, 
boar fossils have been found in Paleolithic and post -Paleolithic settlement 
strata, although not frequently because of the predominantly dry topography 
of these countries. 

In examining the contemporary characteristics of central Asian and 
Caucasian boar — a concentrated succulent vegetal diet taken from the 
soil or subsoil, short legs, relatively poor fur cover and poor adaptation 
to low temperatures and deep snow cover — it is evident that the latter 
features determined the distribution of this animal in the Quaternary. In 
this respect, Formozov's (1946) opinions on the "chionophobic" character 
of boars is perfectly reasonable. Under the varied climatic and topographic 
conditions of a mountain country such as the Caucasus, depth of snow cover, 
335 would not, however, have been such a decisive factor and, in fact, would 
have permitted a satisfactory existence for boar on the Caucasus in the 
Quaternary. 

Settlement strata and burials of the Bronze Age and later cultures on 
the Caucasus almost always contain remains of boars. They are especially 
common in the early settlements of the swampy Colchis near Anaklia and 
in the ancient towns of Kuban. An abundance of boar was characteristic of 
the Middle Ages. This is particularly true for the Don valley where 
remains of these animals comprise as much as 6% of all fossils collected 
in Slav and Khazar strata of Sarkel near Tsimlyanskaya. One can estimate 
to a certain degree the distribution and population of boar in the Caucasus 
during the historical epoch from art objects and literature. There are no 
representations of boars on ornaments and among burial provisions of 
the Bronze Age on the Caucasus; these are noted only from the later 
Scythian epoch. Spectacular boar representations (Figure 130) are found 
in burials of the Kuban area and the Taman Peninsula. 

Moisei Khorenskii (1893 edition) mentions the boar hunts of the rulers 
of the Arsacid dynasty in the Araks valley in the 1st- 2nd century A. D. , 
which places them in ancient Roman time. In this description of the reign 
of ArtavazdIIandthe war with the Romans (Book 2, Chapter 2, page 73) 
he said: "He indulged in food and drink: he wandered and rambled on bogs, 
in reed thickets, on steep slopes, hunting onagers and boars. . . " and further 
(Book 3, Chapter 55): "Another time we had a boar hunt among the burned 
reeds. " 



331 




The decrease in the boar population 
and the restriction of its range was 
more rapid on the Armenian Highland 
because no religious prohibition was 
obtained against boar consumption and 
because the upland slopes were heavily- 
deforested. Boar remains which can 
now be found on the shallowed shores 
of Lake Sevan show how abundant boars 
were on this upland in the past. Hunting 
was the chief operative factor in the 
FIGURE 130. Representation of a boar on a disappearance of boar from [ancient] 

bronze plate fronn the Semibratnoe burial Christian Armenia, a fact that is 

(The Hermitage) pointed up by the contemporary 

existence of a large boar population 
in considerably drier regions of Iran and Turkey. In Georgia the many 
mountain forests and almost inaccessible swamps maintained a thriving 
boar population over a long time, and it was only the 19th and 20th centuries 
that saw the beginning of the rapid extinction of the species. 

In Moslem regions, on the other hand, the development of agriculture, 
especially rice and melon culture, furnished a nutritional base for a 
stabilized boar population. The Lenkoran lowlands and the Samur forests 
are examples. In general, the spread of the Islamic population during the 
Middle Ages on the Caucasus was, perhaps, one of the primary causes 
of the extinction of the range and the population growth during historical 
times. Even now, there are more boars in Azerbaidzhan and Dagestan than 
in other places. One can judge the abundance of boars on the Caucasus in 
the 18th- 19th centuries from the many notations contributed by Caucasian 
naturalists, students of local lore and travelers. 

The rapid decrease in boar population probably commenced in the 1850's 
336 after Russian settlement of the Caucasus. This decline was first observed 
on the lower Don, then on the Kuban and later on the Terek and Sulak. 

Boars were already rare in the area of the Don estuary and on the Azov 
coast by the beginning of the 19th century, and they completely disappeared 
about 1812. 

In the 1870's boars were still found in the forests near Stavropol (Dinnik, 
1914a) from which they have completely disappeared in the 20th century. 

Before 1900 the boar population on the northwestern Caucasus declined 
considerably because of the growth of settlements of Armenian hunters who 
hunted throughout the year. 

A small number of boars inhabited the reed thickets of the Kuban as recently 
as 1932 in the vicinity of the Cossack villages of Grivenskaya and Petrovskaya. 
A local hunter would usually kill no more than ten of these animals per 
season. 

In the 1930's a small boar population was encountered in the alder-grown 
bogs of the lower Rion between Poti and Samtredia — an area which had 
swarmed with these animals over many millennia. In general, western 
Georgia, Imeretia and Abkhazia supported only a small boar population 
by 1930. 

In 1909 in the Karayazy forest east of Tbilisi, as many as 50 boars were 
killed during one hunting day. In 1911, 120 boars were killed in a two-day 



332 



period (Kalinovskii, 1900; Markov, 1931a, 1935). Today it is hardly 
possible to kill five boars in Karayazy in a day. 

In the Kura-Araks Plain the boar population fluctuated throughout the 
Quaternary period as the Caspian Sea level rose and fell and created breaks 
in the coastal ridges which resulted in the displacement of wormwood 
steppes by vast reed — cattail swamps and lakes. 

Such a break of the xA.raks occurred in 1896 when vast floods of Akh-Chala 
occurred on the Mugan steppe (Shelkovnikov, 1907). Even today such 
adverse factors as the height of the Kura in flood, the extent to which the 
steppe lakes fill and the burning-back of reeds largely control fluctuations 
in boar population. 

The booklet "Agriculture and Forest Economy of the Lenkoran District" 
(1914, pp. 21-22) contains the statement: "Ten years ago the number of 
wild pigs was so high that they became a real menace but when boar hunting 
became commercialized, the nunaber of pigs was considerably reduced. 
Every year Armenians from the Elizavetpol Province came here to hunt 
wild pigs for their skins to use in the manufacture of shoes. Local Russian 
hunters kill the swine for their meat. For the last two years, about 
5,000 skins were sent to Elizavetpol Province through Lenkoran. " 

During the 1930's the forests of northeastern Karabakh and the northern 
slopes of the Lesser Caucasus as far as Georgia did not contain boar. 
They were, however, encountered in the Akera-Chai valley and on the 
afforested northern slopes of the Zangezur Range where the local population 
was replenished by boars migrating through the Araks from the juniper 
forests of northern Iran. 

Before the later 1940's, a vast tugai forest extended 70 km along the 
Kura River from the outlets of the lora and the Alazan to Mingechaur; 
it was destroyed preparatory to the flooding of the Mingechaur Reservoir. 
337 The boars which inhabited these tugai forests were either killed or relocated. 
In the 1950's, after the closing of the Mingechaur Dam, the tugai forests 
below Evlakh began to dry out rapidly and, as a consequence, disappeared 
from this area also. 

The decline of the boar population in the last decades is visible even in 
Talysh where the mountain forests are thought to have abounded in boar. 

On the whole, the decline of the boar on the Caucasus during historical 
time is a result of its being hunted, both as a valuable trade animal and 
as a pest. Anthropogenic changes in the landscape and epizootics were of 
secondary importance. 

The present range of boars on the Caucasus is still important. They are 
observed all over the northern slopes of the Main Range and in the foothills, 
even near Kislovodsk andNal'chik. There are few boars in mountainous 
Dagestan. The densest population is in the Caucasian Reservation, in the 
lower reaches of the Terek and the Sulak, in the Alazan-Agrichai lowlands, 
in parts of the Kura and Araks tugai forests and in Talysh (Map 80). In 
adjacent territory of bran, the boar is most numerous on the northern slopes 
of the Elburz Range (Sarkisov, I944e). 

The vertical distribution of boar on the Caucasus extends from sea 
level to the alpine zone. The piedmont lowlands are inhabited more 
continuously; in the mountains distribution depends upon the availability' of 
seasonal forage and the depth of the snow cover. According to the data of 
Donaurov and Teplov (1938), boars are observed in the following zones of 



333 



the Caucasian Reservation (percentage /zone /year): broadleaf forest zone — 
48%; dark coniferous forest zone — 39%; subalpine zone — 11 %; and alpine 
zone — 2 %. 

Table 72 is based upon observations of the author's in 1935-36 in the 
Alazan valley and the Zakataly Reservation; percentages reflect fresh traces 
of boar (crouching imprints, evidence of digging, and excrements) observed 
in each zone or biotope. 



TABLE 72. Incidence of boar tracks observed on the southern slope of the Greater Caucasus in Azerbaidzhan 



Zones and biotopes 


Percentage of total traces observed on prescribed 
20- km route 




August 1935 


October 1935 


January 1936 


Lowland forests of Alazan 


45 
32 
10 
10 
3 


15 
35 
32 
28 


52 


Foothills with fruit trees 


33 


Beech forests of the middle zone 


15 


Oak forest edges of the subalpine zone 




Alpine meadows 









In Talysh and the Lenkoran lowlands, lacking an alpine zone, the 
distribution of boar is somewhat different: in the mountain semidesert zones, 
where grain fields and fruit trees are plentiful near the upper forest edge, 
the boars concentrate in June and July, while in fall and winter they mainly 
inhabit the lowland and the middle forest zone. The daily activity cycle and 
338 distribution of boar are influenced, not only by fodder availability, but by 
the degree and kind of hunting. In eastern Ciscaucasia and Transcaucasia, 
boars often couch in the vicinity of the railroad (Dinnik, 1914a), in gardens 
and arriong weeds growing in borders of cotton fields (in the Shirvan and 
Mugan steppes). 

Where they are not hunted or disturbed, boars commence to feed 
at dawn and, in the reed thickets of the Kyzyl-Agach Reservation, they 
even wander by day. In places where they are constantly harassed, boars 
may feed only late at night, and before daybreak will travel sometimes as 
much as 10-15 km from their feeding ground to find couching places. 

The animal's preferred biotopes in the eastern half of the Caucasian 
Isthmus are impassable thickets of tugai, lower forests of Caucasian olive 
and willow, bramble shrubs and reed thickets. These thickets are the 
primordial biotope suitable to morphological adaptations of the animal. 

Any disruptions in the pattern of their horizontal and vertical migrations 
in search of food or an escape from deep snows cause the death of 
contemporary boars. Dinnik has recorded (1914a, pp. 47-48) episodes of 
mass mortality among boars in the 1880's and between 1902-1908 caused by 
deep snows, cold and hunger. It seems possible to us that this situation 
could have been created by their inability to migrate to the warm Kuban Plain. 

They could have reached the Kuban valley from the region of deep snow of 
the mountain forests through stands of forest or tugai. But by the 20th 
century, their migratory instinct had been suppressed as the result of 



334 



harassment and habitat modifications, such as human settlement and 
the deforestation of the piedmont plain. 

In the 1940's the total boar yield on the Caucasus was 8,000: 
Azerbaidzhan — 4,000; northern Caucasus — 2,000; Dagestan — 1,500; 
Georgia — 300; Armenia — 200 (Vereshchagin, 1947d). 

The future of the boar population of the Caucasus essentially depends 
upon the further development of the agricultural and forest economies. 

Boars are pests in fields of corn, barley, wheat, rice and melons and 
where these crops are grown they will be exterminated or considerably 
reduced. 

In eastern Transcaucasia the development of irrigation flood-control 
and cotton farming will inevitably reduce the boar population through the 
reduction of protective cover. 

If hunting quotas were established boars could exist in the mountain 
forests of the Greater Caucasus and Talysh for an indeterminate time. 



Family CERVIDAE 

Pliocene strata of the Caucasus have been found to contain remains of 
representatives of the genera Eucladocerus, Procapreolus and 
Pliocervus, and Cervus, Dama, Megaceros, Alces and 
Capreolus specimens are known from post-Tertiary deposits. Of these, 
only Cervus elaphus, Alces and Capreolus have survived to 
the present. 

Red deer — С e r vus elaphus L. (s.lato). Fossils of the red deer 
group (C. elaphus) are known in Europe from the Pliocene. From 
339 Mediterranean deposits ten different types of deer are recorded, but only 
from the Pleistocene. In the Caucasus the remains of these deer have been 
found from the Upper Pleistocene to the present. 

For the most part these belong to the extant C. elaphus maral Ogilby. 
The location of the fossil remains (Map 83) indicate the wide distribution 
of this species. 

North of the Isthmus C. elaphus was as common a settler of the 
floodplains and flood valleys in the Pleistocene as the primitive bison and 
the horse. In the Novocherkassk Museum its remains rank fourth after 
elephant, primitive bison and tur fossils. In riverbed deposits found on 
sand banks, bones of C. elaphus account for 3.3% of all the bones 
accumulated on the lower Don and 0.5 to 5.0% on the lower Kama and 
middle Volga. 

South of the Caucasus on the Iranian Plateau remains of Pleistocene 
C. elaphus were found in Paleolithic and post -Paleolithic strata of many 
caves and in diluviuna northeast of Maragheh. It is important to mention 
that in caves in the vicinity of Urmia and Behistun bones of deer account 
for 57 % of the total number of extracted bones. In Mesolithic and Neolithic 
strata of the Belt cave in the eastern extremity of the Elburz Range there 
were few deer remains —only 0.8% (Coon, 1951). In the Neolithic strata 
of the ancient townof Annau in western Turkmenia deer remains are rare 
(Duerst, 1908). C. elaphus was a common food of Paleolithic hunters 
in Lebanon, Syria and Palestine. In spite of its frequent occurrence, 
there can be no doubt that the habitation of deer in southwest Asia was 
confined to river valleys and wooded slopes from at least the Pliocene. 



335 



Fossils of С. elaphus are as numerous in postglacial deposits as 
they are in Pleistocene deposits. In the Holocene loams of the Stavropol 
and Pyatogor'e areas, in the Yegorlyk, Kuman and Podkumok river valleys, and 
in talus fragments of the slopes and banks of Lake Sevan on the Armenian 
Highland, skulls, teeth and antlers are especially common. 

These findings (see Chapter II) indicate that deer was widely distributed 
over the Ciscaucasian steppes and Transcaucasian plateaus, at least in 
the spring and fall and at the beginning of winter. 

This earlier distribution of deer on the steppes should not be associated 
with a wide deforestation of these areas as Shelkovnikov (1930), Dal' (1947a) 
and others thought. In the spring — summer season deer do not require large 
forests; on the contrary, they migrate from the forests to steppe and 
meadow areas to escape from bloodsucking insects. 

At present deer inhabit the Don, Volga and Ural valleys and adjacent 
steppe and forest areas. Doubtless the Caucasian population of C. elaphus 
was always related to the Eastern European population. In the strata of 
the Middle Ages in the Sarkel fortress near Tsimlyanskaya C. elaphus 
bones comprised 4.3% of all the fossils investigated. 

During the post-Paleolithic C. elaphus was the favorite subject of 
painters and sculptors of the Caucasus; it was used in the decoration of 

340 pottery, on arms and on clothes. Beginning with drawings on rocks 

in Kabristan (Figure 13 l), the deer subject is repeated in bronze pendants 
of Kobanian culture, in the ornamentation of Khodzhala and Sevan bronze 
belt buckles and especially in gold articles of Scythian-Sarmatian culture 
of the Trans-Kuban Plain and south Ukraine (Figures 132-133). Deer figured 
importantly in religious belief and rituals. According to legends of the 
Abkhazians and Imeretians, deer coming on the Feast of St. George to 
caves or chapels dedicated to the saint were killed for feasting (Dzhanashiya, 
1915). In a number of Ossetian dzuars antlers and sacrificial skulls of deer 
predominate among the remains of other wild, hoofed mammals. All this 
indicates that the deer population during the postglacial period did not 
decrease as rapidly and catastrophically as did that of tur and bison. 

Osseous remains of deer found in Quaternary deposits of steppe regions 
are often cited by naturalists as evidence of either a former steppe way 
of life or an earlier existence of forests in steppe regions. Recent 
observations of deer behavior and feeding habits and of the structure of 
the lower molars clearly indicate that the C. elaphus of Pleistocene and 
historical time was an inhabitant of the steppes, especially in spring and 
fall. During these seasons they grazed on the young grasses of the steppes, 
and at other times of the year inhabited the floodplains and oak forests 
in ravines. Deer were widely distributed in steppe and forest-steppe at 
least until the late Middle Ages. Guillaume le Vasseur de Beauplan 
(1832 edition, p. 92) noted flocks of deer in the Ukrainian steppes at the 
beginning of the 17th century: "Deer, fallow deer and saiga are 
encountered in flocks in the Ukrainian steppes. " The present existence of 
deer on the steppe -pastures of the Askaniya-Nova Reservation gives 
credibility to this earlier report. 

341 Shal'kovskii (1885, p. 189) also noted in his account of 18th-century 
Zaporozh'e that hunters "also caught hares, wolves and deer which, until 
1760, inhabited the marvellous forests of the Trans-Dnieper Zaporozh'e." 



336 



(340) 









FIGURE 131. Beyuk-Dag rock south of Baku showing representation of a deer hunt 

Photograph by the author, 1946 





FIGURE 132. Kobanian artifacts from North Ossetia (Uvarova, 1900) 
1 — bronze rod representing a deer hunt; 2 — deer head 



337 



During historical time a marked decline in the deer population can be 
traced to the development of cattle herding and agriculture. The greatest 
decline in the deer population of the East European and Caucasian steppe 
and forest-steppe zones, however, probably occurred when horses were 
domesticated and used in stag hunting. During the campaigns of the 
Scythians, Huns, Kipjhaks and later the Mongolians, deer living on the 
steppes were rapidly decimated. It is possible that stag hunts forced the 
retreat of deer into the forests and forested foothills where they were out 
of reach of the mounted hunters. At the beginning of the 15th century, 
Josaf Barbaro (1836 edition) witnessed a Tartar "pagany" — mounted 
hunt — for bustards and other animals which formed an encirclement of 
120 miles (I ) "and [saw] deer and other animals escaping from the Tartars." 

The first phase in the depletion of С . elaphus during historical times 
was the extermination of its steppe population. 

The second phase of decline began in the late Middle Ages as a result 
of agricultural development and the rapid eradication of forests. This forest 
extermination was accompanied by desiccation and subsequent dwindling 
342 of land in the Caucasian foothills and high plateaus. 

Historical data on the distribution and yield of deer have been preserved 
since the 16th century. Georgian and Iranian chronicles and the ancient 
literature of Georgia, Azerbaidzhan and Iran mention the deer hunts. 
Thousands of peasant beaters and warriors took part in hunts organized 
by the Persian Shahs during the late Middle Ages in the forests of Gilan 
and Karabakh. 

The populousness of deer in Mazandaran and Gilan in the 16th century 
can be reckoned from an end-of-the century account by Adam Olearius 
(1870 edition, p. 730): the Shah Takhmasp and his retinue killed 2,000 deer 
and goitered gazelles — "agu" — and he ordered the "building of a tower 
from their antlers. " 




FIGURE 133. Scythian artifact from the Trans-Kuban Plain 

Gold-plated reclining deer from a hill near the Cossack village of 
Kostromskaya 



338 



In "The Geography of Georgia" (1904), Prince Vakhusti reports that, 
at the beginning of the 18th century, the 94th czar, Vakhtang VI, and his 
entourage killed 180 deer in one day on the Trialet ridge southwest of Tiflis. 

The historian Sekhnia Chkhaidze writes in "The History of Georgia" 
(1945, p. 408): "The Czar Vakhtang Levanovich (1703-1721) invited persons 
of high rank to a hunt in Trialetia and Shanbiani. The first day 60 deer and 
countless other animals were killed. " 

At this time deer still inhabited the reed-grown and forested floodplains 
of the Don, Kuban, Manych, Kuma, Kura, Terek and Sulak rivers. 

The disappearance of deer from Ciscaucasia in the 17th and 18th 
centuries can be attributed to the settlement of the country by Cossacks 
and the widespread use of firearms. Dinnik (1914a, p. 97) reports that he 
found only fragments of antlers as evidence that deer existed there 
"60-70 years ago, " i.e. , in the 1850's and 1860's. 

In the environs of Krasnodar and on the right bank of the Kuban, where 
they had been protected, deer still lived up to the early years of this 
century, when they were finally exterminated during the Civil War. 

According to Dinnik (1914a) the northern limit of their range on the 
northwest Caucasus passed through the Cossack villages of Peredovaya, 
Kaladzhinskaya, Makhashevskaya and north of Maikop. The largest deer 
population was observed in the upper reaches of the Belaya, Bolshaya and 
343 Malaya Laba rivers where 20-30 deer could be encountered in one day. In 
winter, deer ranged as far as Novorossiisk and Kabardinka. Later, the 
distribution area shifted southward. In the early 1930's, the northern range 
limit lay in the foothills along the line of the villages of Solenyi, Buguzh, 
Dakhovsk, Samursk, Neftyanaya and Navaginskaya. The range reached its 
western limits in the upper reaches of the Pshish River (Nasimovich, 1936a) 
In the 1940's deer could still be seen east of the Caucasian Reservation 
near the railroad stations of Predgradnaya and Mikoyan-Sakhar. 

Rossikov (1887) tells of hunting deer on the piedmont plain north of 
Nal'chik as late as the 1870's. 




FIGURE 134. Illustration of deer hunt in Digoria from "polati" [sleeping ledges or bunks] in the 
Digorized cave 

Photograph by author, 1947 



339 



The last of the deer in the piedmonts of North Ossetia were exterminated 
(Figure 134) by the 1920's (Vereshchagin and Naniev, 1949). They survived 
for a longer time on the Sunzha tributaries near Gudermes and they are 
still found in tugai forests along the Sulak and Terek on the Dagestan plain. 

In the Dagestan uplands deer were numerous in the gorges of the Avar 
Koisu. Many antlers have been found in burials in deciduous and coniferous 
forests near Antsukh [Bogaz Range] andTlyadal, and in Khevrusetia [eastern 
Georgia] and Tushetia [Andi Koisu region] (Dinnik, 1914a; Maruashvili, 
1955). The deer migrated to Dagestan from the southern slopes of the 
Greater Caucasus only in the summer to escape from bloodsucking flies, 
returning again in October or November to winter on the southern slopes. 
It is the practice of the Dagestanians to hunt them in the passes from 
Dagestan. 

Deer had disappeared from the Samur-Khachmas lowland forests by the 
beginning of this century and, according to old-time residents of the 
village of Kusary, they were gone from the higher forests of the Kuban area 
by the I880's. 

In the Kura lowlands they could still be encountered in the tugai forests 
and reeds of the middle and lower reaches of the Kura River as late as 
the end of the last century. Older residents of the ancient village of Shil'yan 
and of the Shirvan steppe recall deer living along the reed-grown margins 
of the Lake Shil'yan, and a path through the reeds is still called "тага], 
gel" or "maral elu" (deer path). Deer was already rare in Talysh by the 
time of Radde, Satunin (1905b) and Dinnik (1914a). According to the 
Azerbaidzhan Game Inspector, V. V. Vitovich, the last deer in Talysh was 
344 killed in 1912. 

In Karabakh, deer had disappeared from the Dzegam and Shamkhor 
ravines by the 1920's. Melik-Shakhnazarov (1898) reported that in the 1890's 
in Zangezur deer were so abundant that some hunters killed 150-200 head 
a year. Osipov (1898) tells of deer being common in his time in the vicinity 
of Chaikent (south of contemporary Kirovabad). 

In the 1930's and 1940's some deer still survived in the Karayazy forest 
on the Kura River where in the I890's 15-20 specimens had been killed in 
one day. 

In western Georgia, deer were either exterminated or driven into the 
mountains from the lower Rion, Kodor and Ingur by deforestation at the 
start of World War I. 

The present range of C. elaphus is comprised of four or five isolated 
tracts (Map 83), the largest of which is located in the northwest and includes 
the Caucasian and Teberdinskii reservations in the Pshish, Kurdzhips, 
Belaya, Laba, Malyi Zelenchuk, Bolshoi Zelenchuk and Kuban river basins. 
In the summer, most of the deer inhabit open places in the fir forests in 
the interior of the Caucasian Reservation at altitudes of 1,700-2,200 m. 
Their winter habitats are found in sunny places on the slopes and in the 
valleys of the lower fir belt and the beech-forest zone with a snow cover 
of 0.3-0.5 m. After the sharp decrease during the Civil War, the deer 
population commenced to rise again. A 1934 census (Nasimovich, 1936a) 
showed a total stock of 900 head. By the 1940's, there were at least 
3,000 head of deer in the Caucasian Reservation. 

In northeastern Ciscaucasia deer survived in forests and reed-grown 
areas of the lower Terek from the Cossack village of Chervlennaya to 
Kizlyar, and on the Sulak from Khasavyurt to the lower river. 



340 



These deer live the year around in the tugai stands of the lowlands, 
because there are no longer any alternative locations open to them for 
migration: the northern foothills of Dagestan are now deforested and 
populated. 

In western Transcaucasia a small deer population still inhabits the 
forests on the middle course of the Ingur. In the beech forests of the 
Eorzhomi Reservation (on the eastern spurs of the Adzhar-Imeretia 
Range), El-Mar (E. L. Markov, 1939) counted almost 900 head of deer in 
1934. The summer habitat of the Borzhomi deer had previously been 
located on the high, cold, deforested plateaus of the Dzhavakhetia and 
Akhalkalakhi uplands. In eastern Transcaucasia deer are found in the 
Karayazy forest and on the southern slope of the Main Range from the upper 
Alazan to the Gerdyman-Chai. The largest deer population has been 
observed in the Zakataly Reservation in the Belanukh-Chai and Katekh-Chai 
basins; it was estimated in the 1940's at 1,000 head. To the east, in the 
Kutkashen and Ismailly districts, deer are numerous both in the mountains 
and on the plain, and can be encountered near Chukhur-Kabala, Ismailly, 
Kalandzhak, Rushan-Kend and other villages. 

Our own observations made in the 1930's showed the summer habitat 
of deer to be on the southern slopes of ridges in the upper third of the 
beech forest zone and in the subalpine zone. If there were no cattle or 
dogs, they would venture onto the pastures. In the winter they inhabited 
beech forests in the middle altitudes of ridges with southern exposures. 

During winters with a heavy fall of snow deer sometimes appear on the 
345 open hilly steppe South of the Shemakha. This habit and the year-round 

existence of the Sulak population in reeds and tugai explain why deer remains 
were found together with those of ass and saiga in Pleistocene deposits 
on the Apsheron Peninsula. Notwithstanding its earlier history and these 
present variations, the contemporary C. elaphus must be regarded as 
a mountain-forest species, not accustomed to remote migrations to the 
plain. The differences between various deer populations in habits and in 
the directions their migrations take merely testify to the great adaptability 
of the animal and require a correct reading of its contemporary behavior 
as a response of its higher nervous system to human influence. 

During the 1940's poachers on the Caucasus averaged an annual bag of 
200-250 deer (Vereshchagin, 1947d). 

The Caucasus is the chief deer preserve in the U.S.S.R. The subsequent 
fate of the species in this region will depend upon the effectiveness of 
measures taken for their protection. If poaching is controlled, it will 
allow renewed breeding in the forests of Karabakh, Talysh and the Kusary 
sloping plain. In addition, it should be possible to resettle Caucasian deer 
in the future in the floodplain and watershed forests of the Russian Plain. 

Fallow deer — D a m a cf. mesopotamica Brooke. Fossil remains 
of fallow deer similar to the Mediterranean D. dam a L. are recorded 
from Pliocene and Pleistocene deposits in Italy, England and Germany. 

A fragment of mandible and a fragment of antler discovered on the 
Caucasus in Pleistocene alluvium at Adzhi-Eilas near Yerevan are unique 
and can be reported here as D. mesopotamica only conditionally. 

In southern Asia Minor and Palestine fallow deer was the most frequent 
bag of Paleolithic hunters (Bate, 1937). 



341 



Artistic representations of Mediterranean 
fallow deer heads in bronze (Figure 135) were 
found in the Semibratnoe Scythian burials on the 
Taman Peninsula, and also in burials in the 
southern Ukraine. This is a basis for presuming 
that the distribution areas of the two species, 
i.e., Mediterranean and Mesopotamian, 
encompassed the southern Balkan Peninsula, 
the Black Sea area, Asia Minor, Iraq and Iran 
(Map 82). The fallow deer is an inhabitant of 
broadleaf forests and thickets and of the 
undergrowth of Mediterranean river valleys. 
The species is more thermophilic than Cervus 
e 1 ap h u s. 

By the beginning of the 20th century, the 
fallow deer was nearly extinct in its natural 
range; it has since been bred under protected 
conditions throughout Western Europe and north 
Africa (Ellermann and Morrison-Scott, 1951). 
A successful experiment in breeding fallow 

deer was carried out on the Caucasus nearBorzhomi before the revolution 

(see Chapter VI). 

Through future regulation of hunting, the fallow deer can become a 

worthwhile element inthe piedmont shrub-forest zone of the Caucasus. 

It can populate the forests of Stavropol, the Dagestan piedmont, eastern 

Armenia, the Nakhichevan A.S.S.R. and the Nagorno-Karabakh uplands. 




L, I \ 1 1 

FIGURE 135. Bronze head of fal- 
low deer from Semibratnoe 
Scythian burial (The Hermitage) 



346 





FIGURE 136. Golden plates from Scythian burial with representations similar to giant deer. 
Berlin Museum (Bachofen-Echt, 1937) 



Giant deer — Megaceros euryceros Aldr. Upper Pliocene remains 
of this deer have been found throughout Eurasia. 

Inthe Pleistocene the distribution area of M. euryceros included 
the forest -steppe and the steppe zones of Europe and Asia. Its remains 



342 



are especially common in Khazar strata of the Volga valley, where they are 
found on the banks and comprise 2—8% of all the bone material. They are 
also common in Pleistocene strata of Western Europe. On the British 
Isles fossils have been found in caves, diluvium and peat beds. Some 
paleontologists, reasoning from 12th-century drawings and the rather fresh 
remains found in Irish peat beds, view M. euryceros as an inhabitant 
of Ireland up to the time of the present era (Mitchell and Parkes, 1949). The 
European continent lacks such recent remains indicating that the species 
was probably extinct here in the Upper Pleistocene. However, Pidoplichko's 
report (1951) of a find of M. euryceros remains in the U. S. S. R. , which, 
according to the bone calcination, showed a Recent origin, deserves further 
attention and verification. 

Kitchen middens containing M. euryceros remains, which are common 
in Paleolithic caves of the Crimea, are completely unknown from the caves 
of Palestine, Syria and Lebanon. On the Caucasus Isthmus, fossils have 
been found from the Upper Pliocene — on the Taman Peninsula, through 
the Mousterian — in Abkhazian caves and camp sites on the Trans- Kuban 
Plain (Map 82). M. euryceros inhabited Abkhazia in a mountain 
mesophytic environment. 

The species is rarely found in Pleistocene strata of the eastern 
Transcaucasian steppes, and in the Binagady asphalts only one bone among 
tens of thousands was identified as M. euryceros. No remains of giant 
deer have been found in Holocene deposits, even those as large as the Lake 
Sevan shore and the Sarkel strata of the Middle Ages. It is not depicted in 
any Caucasian Paleolithic or post -Paleolithic drawings and scuptures. 
Stylized representations ofCervus elaphus in gold plate taken from 
Scythian burials on the Trans-Kuban, which somewhat resemble representa- 
347 tionsof M. euryceros (Figures 136, 137), led Bachofen-Echt (1937) to 
the erroneous assumption that giant deer existed on the Caucasus in the 
first millennium B.C. 




FIGURE 137. Skull of M egaceros euryceros from the environs of Rostov-on-Don 

The usual osseous deposits containing M. euryceros found in flood- 
plains and peat beds reveal an ecological similarity to the elk. The 
structure of metapodia and especially of hoof phalanges place the species 
in a position intermediate between elk and ox. It could probably endure 



343 



a deep snow cover, but could not undertake long marches in the snow like 
the elk. It was an inhabitant of the cold forest-steppe and the river valleys 
of the plains, living in floodplain forests and thickets together with 
Elasmotherium and Elaphus trogontherii. 

Caucasian elk — Alces alces caucasicus N. Ver. Remains of 
elk of the A. alces L. group are recorded from Pleistocene deposits of 
the northern Holarctic, but are rarely reported south of the 50th parallel. 
In the western Mediterranean Pleistocene deposits seldom contain 
elk, although other species of the Alces genus are sometimes so identified 
(Wolf, 1938; Pidoplichko, 1951 — map, p. 79, reports all findings as 
A. alces). Earlier reports on finds of elk remains in the Paleolithic of 
Palestine, Syria and Lebanon (Blanckenhorn, 1910) were in error (Picard, 
1937). The same may be said more emphatically for Egypt. North of the 
Caucasus remains of Pleistocene elk have been found at Borshevo II, the 
encampment site near Voronezh, and on scarps of the Volga bank near the 
village of Osypnoi Bugor beyond Astrakhan. The species does not occur in the 
Crimean Paleolithic. On the banks of the middle Volga and especially of 
the lower Kuma, elk remains generally account for 5% of all the large - 
species bones. 

On the Caucasian Isthmus elk remains are recorded for the Upper 
Pleistocene (Upper Paleolithic) from cave strata in western Transcaucasia 
from Khosta to Kutaisi. 

The Mousterian strata of the Akhshtyrskaya cave, the Il'skaya 
encampment and the Apsheron Peninsula asphalts revealed no remains 
of elk. It would be justifiable to conclude that the elk first penetrated the 
Caucasus from the north in the Upper Pleistocene, although it is equally 
possible that the development of the ancestral form could have taken place 
on the Caucasus, or, for that matter, anywhere in the eastern 
Mediterranean. In either case, it is a matter of established fact that the 
elk survived the Upper Pleistocene on the Caucasus and became extinct 
only at the beginning of the 19th century. 

The presence of elk on the Caucasus during the Bronze Age can be 
assumed from bronze buckles and pendants of Kobanian workmanship found 
in archaeological excavations in North Ossetia (Uvarova, 1900, p. 80, 
348 Figure 75; Vereshchagin, 1949a). In the collections of the Moscow State 
Historical Museum there is a bronze buckle with a design of two elk heads 
which was excavated near the village of Atage in the Sunzha valley and 
belongs to the second period of Koban culture (Figure 138). An interesting 
relief resembling an elk is carved on a stone from Euyuk, a Hittite town 
which existed in the first half of the second millennium B.C. in northeastern 
Asia Minor. This suggests the possibility that south of the Caucasus elk 
inhabited forested tracts in Asia Minor in historical time. 

Representations of elk appear in gold and bronze (Figures 138, 139) and 
on the ornamentation of quivers found in burials on the Trans -Kuban 
piedmont plain and in the southern Ukraine. 

Finally, Keller (1909) reported the find of gold pendants depicting elk 
in ancient Greek settlements on the northern shore of the Black Sea. 

All the elk representations of the Scythian epoch are of great importance, 
not only in reconstructing the early range of the elk, but also in as signing the 
origin of gold and bronze artifacts from Scythian burials with ornamental 
and relief designs of animal subjects. Taking into account that the southern 



344 



349 




margin of the elk range in the Holocene passed through the Caucasus and 
the northern Black Sea area, it becomes apparent that most of these 
Scythian artifacts were in fact produced at their cultural source — in 
Scythia — rather than, as often assumed by historians and archaeologists, 
in the southern Balkans, Asia Minor or the Aegean Islands. 

There are references to Caucasian and 
southern European elk in the literature. 

Jean de Luc (1879 edition) in his 
description of a journey nto the Nogai Tartar 
territory at the beginning of tne 17th century 
tells of seeing elks, in addition to wild horses, 
deer and other animals. Clarke's reference 
(1810, p. 386) to two young and very tame 
"elks" which he saw near Ek.'terinodar should 
certainly be assigned to deer rather than to 
elk, and Kirikov was wrong to use this citation 
(1952) as proof of the earlier existence of elk 
on the Caucasus. 

Pallas (1831) reported in his "Zoography" 
that elk was found on the Caucasus, and 
Lul'e (187 j;, in his description of the beliefs 
and customs of the Circassians, wrote of 
Mezitkh, their god of the forests, who rode 
on a boar with golden bristles. At his nod, 
"deer and elks assembled in the forests, and 
then young maidens milked the does. " 

The presence of elk in the contemporary 
fauna of the Caucasus was later denied, 
however, by the best zoologists (Dinnik, 1896, 
1914a; Satunin, 1903b). Satunin wrote: 
"Thus he (Pallas) placed even the elk (A 1 с e s 
alces L. ) on the Caucasus, although there 
can be no doubt that this species never 
inhabited the Caucasus. " In our time, Kulagin 
(1932) and Buturlin (1934) expressed doubts 
about the occurrence of elk on the Caucasus. 
Kulagin's reservations, which he bases on 
Polferov's story, "Eznos", are really only 
tenuously connected with it. In fact, Polferov 
mentions encounters with elk and capercaillie 
somewhere in the taiga zone, and not on the 
Turkish border as Kulagin surmised. 

In the 1930's elk bones were found in 
archaeological excavations of 8th— 12th-century 
towns near the Cossack village of Tsimlyanskaya on the lower Don. 
Figure 140 shows a semi-fossilized skull of ancient elk identified by the 
author in the Stavropol Museum collections in 1944. While investigating 
Ossetian dzuars in the Urukh Basin in 1947, we found seven elk skulls 
(Figure 141). These findings establish that the elk inhabited the foothills 
of the northern Caucasus contemporaneously with the primitive bison, 
both becoming extinct at the beginning of the 18th century (Vereshchagin 



FIGURE 138. Bronze buckle with 
elk head design from the village of 
Atage on the Sunzha (Moscow State 
Historical Museum ) 




FIGURE 139. Bronze plate 
depicting an elk's head (1:2) 
from the southern Ukraine 



345 



and Semenov-Tyan-Shanskii, 1948; 
Vereshchagin and Ganiev, 1949). 



Vereshchagin, 1949a, 1949b, 1955; 




1 1 1 1 1 1 1 II 1 1 



FIGURE 140. Semi- fossilized skull of elk in the Stavropol Museum 

Photograph by author, 1945 

Ritual collections containing remains of elks, similar to those of Digoria, 
are probably to be found in other gorges of the northern slopes of the 
Greater Caucasus, particularly on the Balkar-Cherek. 
350 Even during the time when the Ossetian dzuars of Digoria flourished and 
skulls and horns were being accumulated in them in quantity, elk was 
already rare by comparison with deer and even bison. The seven skulls of 
elk found in our investigation of 16 dzuars were in a 10% ratio to bison 
skulls and 0.8% to deer skulls. The extinction of the species was brought 
about through direct extermination and through the general destruction of 
forests on the Terek-Sunzha plain. It is possible that the last of the elks 
survived until the beginning of the 19th century. 

During the 1950-1951 excavation of the Sarkel fortress on the lower Don 
carried out by The Hermitage (Artamonov, 1952), kitchen middens were 
found to contain many remains of elk, together with remains of deer, boar, 
saiga, bear and beaver. The elk remains comprised 0.3-0.8% of all the 
bones of domesticated and wild animals found during the excavations. * 

Elk remains were also found recently, together with those of reindeer, in the 
Roman strata of Olvia (Pidoplichko and Topachevskii, 1953). 

A. A. Sadovskii has reported (through a verbal communication) his 
observation in 1937 of a mandible of elk among a large quantity of wild 
ungulate bones, including deer and roe deer, which had been uncovered 
during the excavation of the ancient burial of Naokhvami in Mingrelia. 
The relict alder swamps in the lower reaches of the Rion with thickets 
of osier and buckbean are even today an ideal elk habitat. 



K.B. Yur'ev, G.V. Khrabrov and others investigated 200,000 bone fragments under my direction in Sarkel 
in 1952-1953. 



346 



351 




FIGURE 141. Skull of elk from the 
Digorized cave 

Photograph by author, 1948 



The recent existence of elk in the 
Trans -Kuban was confirmed by the find 
of a horn (Figure 142) in the pebbles of 
the Urup River near the Cossack village 
of Otradnaya. The horn is preserved in 
the Krasnodar Museum; although wind- 
eroded on the surface, the inside shows 
it to be quite recent. 

If the elk settled on the Caucasus from 
the north during the Upper Pleistocene, 
it can be assumed that the principal 
penetration route of the species passed 
southward through the valleys of the Don 
and the Volga. However, the possibility 
of an immigration route from the southwest 
through the Balkans and Asia Minor cannot 
be excluded. The distribution of elk on 
the Caucasus was probably limited by the 
foothills (Map 84). Their chosen biotopes 
(Figure 143) were along spring and swamp 
rivers of the Trans -Kuban andTerek-Sunzha 
plains, on the littoral terraces of Abkhazia 
and in the lowlands of Colchis. As 
opposed to C, elaphus, the elk did not 
adapt to the Caucasian Range and to the 
mountain -forest environment. 

When the lumber and hunting industries 
of the country are regulated, it will be 
possible to resettle elk on the Trans -Kuban 
sloping plain in the valleys of the Afips, 
Psekups, Laba and Zelenchuk rivers. 
It is also possible that the species might 
be successfully settled in young mountain 
forests of pine, aspen and willow in the 
Teberda and Caucasian reservations. 

Roe deer — Capreolus capreolus 
L. and C. capreolus pygargus Pall. 
The genus Capreolus is of Middle 
Pliocene origin (Simpson, 1945). The 
ancestor of the contemporary genus is 
which is known from the Lower Pliocene of 



probably Procapreolus 

Europe and Asia. Flerov (1952, pp. 98-99) considers that the contemporary 

roe deer is the descendent of Mio-Pliocene Cervulinae. 

The infrequently found remains of small European roe deer derive from 
various types of Quaternary deposits in Europe and Asia, almost all of 
them within the limits of the contemporary range (Wolf, 1938; Pidoplichko, 
1951). For the Caucasus, roe deer remains are recorded from Upper 
Quaternary travertines of Mount Mashuk and from Transcaucasian caye 
strata, Acheulean to Recent, at Kudaro II, Akhshtyrskaya and Gvardzhilas 
caves and others. A large form of C. pygargus has been recorded from 
further south for Palestine, Syria and Lebanon from Acheulean to Neolithic 
cave strata (Picard, 1937; Bate, 1937). 



347 




FIGURE 142. Horns of elks: 

1 - from the strata of the Sarkel fortress on the Don; 2 — from the pebbles of the Urup River 



On the Russian Plain remains of a large roe deer were found in Upper 
352 Paleolithic strata of Kostenki on the Don and in Upper Quaternary alluvium 
of the middle Volga region (Khryashchevka). To the west, in the Ukraine and 
the Crimea, only remains of smaller specimens were discovered in 
Paleolithic strata. Roe deer remains are numerous in Holocene deposits 
of the Caucasus, e. g. , in Copper— Bronze -Age strata of camp sites near 
Anaklia, in post-Paleolithic strata of the Akhshtyrskaya cave, etc. (Map 85] 

During the excavations of the Sarkel fortress on the Don remains of the 
large Siberian roe deer were found in 8th— 13th-century strata. The North 
Ossetian dzuars, e.g., the Digorized cave, contain preserved skulls and 
horns of small roe deer ofC. capreolus type hunted in Urukh between 
the 15th and 20th centuries. 

Roe deer is seldom depicted on artifacts left by the ancient Caucasian 
tribes. One of the rare examples is a silver bucket from a Bronze Age 
burial ontheTrialet ridge which bears a contour representation of a roe 
deer wounded in the chest by arrows (Kuftin, 1941). 

The range of roe deer on the Caucasus in the Recent has continued to be 
widespread, and today the species is better preserved than other ungulates. 
A larger form, similar to the Siberian, inhabits the northern Caucasus 
(Dinnik, 1914a), and a small form is found in Transcaucasia. The roe deer 
does not occur in the reedlands of the lower Don and Kuban. It is 



1704 



348 



certain that, as late as the Middle Ages, the Caucasian range of the large 
roe deer was connected with the range of the large Siberian roe deer along 
the Don valley. The presence of the large roe deer in Ciscaucasia can 
probably be explained by a relatively recent postglacial migration fronn 
the north. 




FIGURE 143. Caucasian elk 

The ultimate disappearance of roe deer from the island forests near 
Stavropol took place in the 1920's. There are few of these animals 
remaining in the broadleaf forests of the Trans -Kuban sloping plain between 
353 Khadyzhensk and Maikop. But on the northwestern spurs of the Greater 
Caucasus, they are still populous in beech— hornbeam forests (e.g., near 
the Goryachi Klyuch and Krymskaya). 



TABLE 73. Distribution of roe deer in the Zakataly Reservation based on number of tracks encountered 
daily in summer and winter 



Date 


Southern spurs; beech- 
hornbeam and aspen 
forests with glades; 
700-1,800 m 


Deep gullies in 

ravines; beech forests; 

1,200-1,600 m 


Upper forest belt and 

subalpine meadows; 

2,000-2,400 m 


22-26 August 1935 

21- 26 November 1935 (snow: 

15-30 cm) 
1 January 1936 (snow: 

30-50 cm and more) 


8 

16 
7 


2 
3 
1 


1 



349 



In Kabardino-Balkaria and North Ossetia, roe deer inhabit the forests 
of the Chernye Gory, the canyons of the Skalistyi ridge and the sparse 
forests on the northern slopes of the Terek Range. They are not found in 
the Sunzha valley, but have been observed in the tugai along the Terek and 
Sulak between the Cossack villages of Chervlenaya and Kizlyar. The forests 
of the piedmont spurs of the Grozny Region contain a roe deer population, 
but the area near Makhachkala does not, and the many wolves in the 
reedlands of the lower Sulak and Terek preclude the habitation of roe deer 
there. Neither are they numerous in the forests of the Kusary sloping plain. 
On the whole, the distribution of roe deer on the northern slopes of the 
Caucasus reaches its summer limits at altitudes of 1,500-2,000 m. In the 
winters when the snowfall reaches disastrous proportions the animals 
descend to the piedmont plains, sometimes even venturing into the villages 
of the Maikop area (Olenich-Gnenenko, 1955). 

In western Transcaucasia, the population of roe deer is sparse in the 
lower forest belt because of the dense network of villages; it is heavier in 
the middle third of the forest zone. The snowfall in the upper Svanetian 
gorges, e.g., on the Ingur, is too heavy to support the existence of roe 
deer there. 

A small population inhabits the Colchis alder swamps near the Chaladidi 
station and on the middle course of the Pichora River, where Poti hunters 
kill as many as 50 animals yearly. Roe deer is common on the northern 
slopes of the Adzhar-Guri Range and the Trialet ridge. On the southwest, 
its range extends to Asia Minor. In Palestine roe deer was still common 
in the last century in the Carmel Mountains where the last specimen was 
killed in 1912 (Bodenheimer, 1935). 

In east Transcaucasia it is found on the southern slopes of the Main 
Range inhabiting mainly the lower third of the forest belt where there are 
thickets of Cornelian cherry, hawthorn and hornbeam, or the glades of 
beech— hornbeam forests overgrown with giant fern and blackberry. 

In 1935 we made the following observations of fresh roe deer tracks on 
rivulet shoals and in the snow in the Zakataly Reservation as they were 
distributed altitudinally along the Tala-Chai and the Katekh-Chai. (Table 73). 
354 On the Alazan-Avtaran lowlands the habitat of roe deer is confined to 

the oak— Caucasian-wing-nut forests, and the population is greater in winter 
than in summer. The species does not inhabit the tugai, reedlands and 
oases of the Kura- Araks lowlands. 

On the Lesser Caucasus roe deer can be found wherever there are large 
stands of deciduous forest. In summer it inhabits subalpine tall-grass 
meadows near relict forest islands at altitudes of 1,800 and 2,000 m in the 
gorges of the upper Debed and Akstafa rivers. 

In Karabakh many roe deer are to be found in the Terter-Chai beech 
forests. They are not encountered on the southern slopes of the Armenian 
Highland. 

In Talysh and in the forests of the El'brus Range, the few deer live in 
isolated colonies in well-defined areas of some of the gorges. Only single 
tracks could be observed in the Vassaru-Chai and Vilyash-Chai ravines in 
our investigations during June and July of 1945. 

Broadly stated, roe deer is frequently found on the Caucasus, but never 
in the numbers in which it is found in the piedmonts of northeastern Tien- 
Shan or the Maritime Territory. 



350 



In the Iranian uplands, roe deer are sparsely settled in relict forest 
islands. According to Blanford (1876) roe deer inhabited only the forests 
of Asterabad. The southern margin of the species range probably 
lies in the passes of northern Mesopotamia. 

Although the principal range of roe deer is now located in northern 
Eurasia, it can certainly be regarded as an Upper Pliocene species of the 
Caucasus whose closest ancestors are to be found in the Lower Pliocene 
of Transcaucasia and throughout the Mediterranean. 

The roe deer is adapted to a forest habitat and a diet comprised mainly 
of leaves; it has the capacity for very swift movement and high jumps in 
shrubs and tall grass; it does not endure deep snow. 

The rapid decline of the roe deer population and range on the Caucasus 
occurred in the last century as a result of unrestricted hunting and 
deforestation (see Chapter VI). 

The present yield of roe deer is estimated as follows: northern Caucasus 
— 4,600; Azerbaidzhan — 1,500; Georgia — 700; Armenia — 300; Dagestan — 
200 (Vereshchagin, 1947d). 

The development of agricultural shelterbelts in the Ciscaucasian plain 
will undoubtedly create ample opportunities for the breeding of roe deer 
in a cultivated landscape. 



Family BOVIDAE 

The Quaternary fauna of the Caucasian Isthmus includes representatives 
of the sub-families Bovinae, Antilopinae and Caprinae. Seven genera occur 
in the Holocene fauna: Gazella, Saiga, Rupicapra, Capra, 
О vis. Bos and Bison. 

Goitered gazelle — G a z e 1 1 a subgutturosa Guld. Fossil remains 
of a number of Gazella species are known from the Upper Miocene- Lower 
Pliocene. The genus is generally regarded as a Miocene species. Fragments 
of horn stems and teeth of gazelles, which show some similarity 
to the Mediterranean Pliocene G. deperdita Gerv., have been found on 
the Caucasus in Sarmatian deposits near Georgievsk, in Middle Pliocene 
sandstones south of Baku and in Lower Quaternary strata of the Taman 
Peninsula. 
355 In addition, a fragment of horn core which resembles goitered gazelle 
horn was found in postglacial loams in the upper reaches of the Kalaus, 
southwest of Stavropol (Ryabinin, 1918). 

Remains which are true to goitered gazelle type were reported for 
Pleistocene strata of the Tamtama and Bisotun caves in central Iran (Coon's 
excavations, 1951), for Mesolithic strata of caves near Dzhebel and 
Krasnovodsk (Okladnikov's excavations, 1949) and for Anau strata near 
Ashkhabad (Duerst, 1908). 

Picard (1937) and Bate (1937) have recorded remains of gazelle similar 
to G. gazella Pall., G. dorcas L. and G. subgutturosa from 
Paleolithic cave strata (Acheulean to Upper Neolithic) of Lebanon, Syria 
and Palestine. No fossil remains of goitered gazelle have been found in 
Pleistocene deposits of the Russian Plain nor in the Middle and Upper 
Pleistocene of eastern Transcaucasia. Validated finds of goitered gazelle 
remains on the Caucasus are recorded only from Holocene strata of the 
historical age in Ciscaucasia and Transcaucasia (Map 87). 



351 



Goitered gazelle remains are particularly numerous in the strata of 
ancient Baku in which they account for 14.5% of all the bones found. Their 
ratio to sheep and goat bones is 3.9:10. 

Most of the present range of goitered gazelle lies within northwestern 
Iran and Afghanistan and the deserts of central Asia. 

The absence of goitered gazelle in the Binagady fauna of the Middle 
Pleistocene and the frequency of its occurrence in Holocene strata of the 
eastern part of the Isthmus lead to the conclusion that the species appeared 
relatively late in postglacial time replacing the saiga in Azerbaidzhan. 
It probably penetrated directly from the south through the ranges of 
northwestern Iran which border the Mugan steppe on the south. This 
extension of the range would have been possible in the dry, warm phase of 
the Holocene. There is also the possibility that the species was imported 
to the Kura lowlands by ancient nomadic tribes which spent the summer 
in the northern Iran uplands and the winter on the Mugan steppe. 

Cattle herders on the Mil'skaya and Mugan steppes capture young goitered 
gazelles which are easily tamed and tether them close to their tents. 
In Syria, Palestine and Lebanon and throughout the Near and Middle East 
newborn gazelles were caught each year by the thousands for domestication 
or for their meat. Usamah ibn Munkidh (1922-23 edition, p. 204) described 
the hunting of gazelles: "When young gazelles are born, the hunters go on 
foot and take the young which were born that night, the previous night or 
two or three nights before. In one day, 3,000 young were caught near the 
Dzhabar fortress. " Young adult gazelles were sent as gifts to chiefs and 
princes, sometimes over very long distances. 

The pursuit of goitered gazelles by Median, Albanian, Persian, Arabian 
and Mongolian horsemen was probably one of the causes of the separation 
of the species range on the plateaus and its penetration into isolated valleys. 
Even now when gazelles are hunted on horseback or in automobiles over 
long distances, it can be observed that they will seek refuge in solonchak, 
tugai and reed thickets or escape into creek valleys or onto the slopes of 
clayey ridges. From these places they emerge one or days later onto the 
open plain because they are frequently vulnerable to attack by wolves in the 
thickets and mountains. 

Ancient writings give evidence on the large population and area of 
distribution of goitered gazelle in eastern and southern Transcaucasia in 
the Middle Ages. Nizami Ganjawi (l3th century A. D. ) frequently employed 
the goitered gazelle as a poetic metaphor. 

Olearius (1870 edition) mentions herds of goitered gazelles which he 
observed in 1633 near Dzhevat [near present-day Sabirabad] on the Mugan 
Steppe and commented on the abundance of these animals on the adjacent 
Iranian Plateau. In the 18th century goitered gazelle existed on Svyatoi 
Island (now Artem Island) which they had reached either through importation 
by Arabs for breeding purposes or in a migration from the Apsheron 
Peninsula during some postglacial drop in the Caspian Sea level. They were 
observed there in 1720-1724 by Soimonov who wrote (1763, p. 110): 
"However, the island is inhabited by many wild goats which reproduce there 
and it is not known what their food is if not mosses growing on stones. " 
. . . "December 3 (1724). . . whereas on Svyatoi Island Don Cossacks with 
their general killed some wild goats. " 

Academician Dorn (1875) noted that Svyatoi Island was called 
by the Arabs "the island of steppe sheep. " Goitered gazelles were observed 



352 



in the Araks valley in the middle of the last century by Nordman (1840) and 
Chopin (1852). They were exterminated in this area only in the second half 
of the century through game drives. 

The species is adapted to a habitat on level semidesert with a solid 
footing (for details, see Vereshchagin, 1937). The animal's great speed 
(up to 50-55 km/hr) permits easy escape from wolves and even cheetahs, 
except on a boggy soil. It is not found, therefore, on newly plowed or 
especially on irrigated soils . Goitered gazelle does inhabit barley and wheat 
fields on the Adzhinour Plateau, but only those that are not irrigated and 
whose soil is more or less compacted. 

While the goitered gazelle is more numerous on level steppes, it is also 
encountered in the Tertiary hills of Dzheiran-Chel, Boz-Dag, Turut-Sarudzha 
and Kabristan — areas of rugged relief with mountain xerophytes. Deep 
frosts and heavy snowfalls bring mass mortality to goitered gazelle 
inhabiting the Kura lowlands (Vereshchagin and Dyunin, 1949). In general, 
the goitered gazelle is a chinophobe because it cannot reach food under the 
snow nor move rapidly over a deep snow cover. This characteristic also 
confirms its recent migration to Transcaucasia from the south. 

Fluctuations in the Caspian Sea level, breaches of the coastal ridges and 
the flooding of the Kura and Araks rivers, as well as changes in the nomadic 
and settled populations of eastern Transcaucasia and in their agricultural 
techniques, have produced far-reaching changes in the range and population 
of goitered gazelle in this area in historical time. 

In the 1920's, the distribution area of goitered gazelle coincided with the 
saltwort solonchak-semidesert and andropogon steppes, and reached nearly 
to Tiflis on the west. 

The sharp population decline and the shrinkage of range in eastern 
Transcaucasia began in the 1920's and accelerated in the 1930's when the 
357 development of motor and air transport made it possible for poachers to 
work mass destruction on the species. It was driven from the Apsheron 
Peninsula early in the century and from the Shirak steppe in the 1920's. 

In the 1930's hunting with cars and trucks developed to such an extent 
that the goitered gazelle was nearly exterminated in the northern part of 
the Shirvan steppe. Just in the area around the Kara-Su railroad station 
poachers bagged more than 500 goitered gazelles annually. Poachers from 
the cotton sovkhozes of the Mil'skaya steppe bagged from 500 to 600 gazelles 
each year with the use of automobiles, and this steppe yielded the same 
number to herdsmen wintering there . In 1937 goitered gazelle could be 
encountered in eight isolated areas of semidesert and poorly-developed tracts 
of the Kura steppes and the ridges of Adzhinour and Kabristan comprising 
approximately 8,219 km^. Its population was in the 5,000-6,000 range 
(Vereshchagin, 1939a) (Map 87). During the years that followed, the range 
and the number of animals decreased rapidly as poaching techniques were 
motorized and as semidesert land was developed for cotton growing. 

In the 1940's the species was displaced from the territory north of the 
Sumgait River. In Kabristan it was confined to deep semidesert ridges by 
the development of the oil industry. Small herds of goitered gazelle could 
sometimes be encountered in 1935-1937 on the Gezdek Plateau and near 
Mount Kergez 10 km southwest of Baku. By the 1940's only single 
specimens could be observed and those no closer than 25-30 km from Baku 
in the Kara Dag and Shakhi-Kai region. 



353 



With the irrigation and plowing of the Kura-Araks lowlands the goitered 
gazelle will become extinct in eastern Transcaucasia Viless a steppe 
preserve is established and effective measures taken for the protection of 
this animal. 

Saiga — Saiga tatarica L. Saiga remains are recorded for 
Pleistocene and Holocene strata from the British Isles in the west to the 
Novosibirskie Islands in the north and Alaska in the east (Nehring, 1890; 
Cherskii, 1891; V.I. Gromov, 1948; Pidoplichko, 1951). The stability 
of the adaptive features of the saiga from the Middle Pleistocene to the 
present points toward a Pliocene or even Miocene origin for the genus. 
The lack of findings of closely related forms in Pliocene strata of the 
Palaearctic is rather mysterious, but, in view of the adaptation of the 
contemporary species, it can be assumed to have developed in steppe and 
savannah environments. 

Although the Pleistocene fauna of the northern shores of the Mediterranean 
generally resembles that of the middle belt of Europe, saiga remains have 
not been found in southern France, Italy and Greece (Wolf, 1938, 1939). 
The most southerly finds of Pleistocene saiga remains have been in the 
Crimean steppes and on the Caucasian Isthmus. On the Russian Plain saiga 
bones present massive and certain evidence of Middle and Upper Pleistocene 
deposits. 

On the Caucasian Isthmus fossil remains of saiga have been reported 
from Ciscaucasia and eastern Transcaucasia (Map 86). 

There has probably been no penetration of saiga to the south of 
Transcaucasia. Nor has it been found in western Transcaucasia which also 
has long been a woodless area of rugged relief. 

The appearance of the species on the Caucasus can be dated at least 
Middle Pleistocene. 
358 The saiga remains of at least 70 specimens found in the Binagady 

bituminous strata of the Apsheron Peninsula are convincing evidence that 
the penetration route of European -Asian steppe forms from the north 
followed the western coast of the Caspian Sea. In the pre-Khazar stage 
of the ancient Caspian regression, the ecological barrier which the 
Khachmas lowland forests of the Kusary sloping plain now present to steppe 
species probably did not exist. The winter migrations of saiga recorded 
by Glitsch (1865) raise questions on the stability of saiga habitats in eastern 
Transcaucasia. Is it possible that these were only seasonal migrations 
taken during cold winters? The answer is supplied in a study of the age 
groups of animals which perished in the bitumen, based on 60 mandibles 
with preserved molars: 

1. Young — June- July; milk molars worn down; first true molar erupted: 
33.3%. 

2. Young — August-September; first true molar erupted and worn down: 
21.7%. 

3. Semi-adult —November; second true molar erupted; milk teeth 
replaced by permanent teeth: 5.0%. 

4. Adults and aged specimens —worn-down true molars: 40%. 
This data shows beyond doubt that saiga reproduced on the Apsheron 

and inhabited eastern Transcaucasia at least in summer, in fall and at the 
start of winter. This habitation continued until some time later, and was 
probably terminated only by the Khvalynsk transgression which greatly 
constricted the developing steppe belt of the piedmont. 



354 



Saiga inhabited the Trans-Kuban piedmont plain and the Taman Peninsula 
of Ciscaucasia in the Upper Pleistocene together with primitive bison and 
giant deer. But whether these habitations represented only winter migrations 
from the Kuban valley and through the Kerch Strait from Crimea is difficult 
to judge from the existing data. The fact that such migrations were possible 
is confirmed by the run of saiga on the Aral Sea islands (Berg, 1905). 

In postglacial time the saiga range was still enormous, extending from 
the Carpathians to Mongolia. Saiga inhabited the Ciscaucasian plain during 
the Holocene from the Azov to the Caspian coast. From this area and 
particularly from adjacent areas of the Russian Plain multiple findings of 
Holocene remains of saiga have been taken. They have been found in cultural 
strata in the following proportions (ratio to total number of bones extracted 
at each site expressed in percentages): l) Cossack village of Tsimlyanskaya, 
Bronze Age, 15th- 10th centuries B.C. — 5 % in one excavation area, 
12% in another; 2) the farmstead of Krasnyi Yar near Tsimlyanskaya, 
Late Bronze Age — 4.2%; 3) the village of Lugovoe in the Assa valley, 
Scythian strata — one horn among a great number of remains of domestic 
animals; 4) Cossack village of Tsimlyanskaya, Khazar strata, 8th- 10th 
centuries A. D. —14.8%; 5) Sarkel fortress near Tsimlyanskaya, 9th-llth 
centuries A. D. , from 1934 excavations — 6.5%, and from 1950 excavations — 
3.3%. 

The saiga is not depicted in drawings, sculptures or folklore of the 
Caucasian peoples, except for a representation of antelope on a vase from 
the Maikop burial (Figure 128) and unrealistic recent drawings that are 
almost contemporary reported from Chokrak sandstones near Kapchugai 
in Dagestan by Markovin (1953). 
359 In the Ukraine saiga was common in the later Middle Ages (Beauplan, 
1823 edition). Litvin (1890 edition, p. 48) recorded saiga and roe deer in 
the province of Kiev of the Lithuanian Grand Duchy as follows: "Wild goats 
run across from the steppes into the forests in winter and from the forests 
to the steppes in summer in such large numbers that each peasant kills up 
to 2,000 of them each year. " This was approximately the year 1550. 

In the last century the southern range limit of the species in Trans- 
caucasia lay alongthe Ust'-Urt, the Aral and the northern Kyzyl-Kum. 

During the 19th century and at the beginning of the 20th, the decline in 
saiga population on all surviving ranges proceeded very rapidly as the result 
of direct extermination (Glitsch, 1865; Dinnik, 1914a; and others). The 
plowing of land was also a contributing factor. At the time of Rossikov 
(1887) saiga still inhabited the lower reaches of the Malka, and in the 1890's 
could still be encountered near Stavropol: Dinnik (1914a) reports their 
being killed near the Sengileevskoe Lake and the village of Staro-Mar 'evka. 
At his time saiga were common near Arzgir, Raguli and Turkmenskaya 
Stavka [Letnyaya Stavka, Stavropol Territory] but had become extinct 
in the vicinity of Bol'shaya Kugul'ta, Dzhalga, Burukshum and Takhta. 
By the beginning of the 20th century they were no longer to be 
found in the western Manych area (Bogachev, 1918). At this time numerous 
armed detachments on the steppes contributed to the disastrous decrease 
in saiga population. They survived only in certain parts of the Chernye 
Zemli, in the lower reaches of the Kuma, Volga, Ural and Emba rivers, 
in northern Kazakhstan and in Mongolia. They completely disappeared 
• from the Terek-Kuma semidesert. In the 1930's migrating herds were 



355 



seldom seen near Arzgir and Terekli-Mekteb. The establishment of hunting 
limits, the confiscation of rifled arms and the gradual depopulation of the 
semidesert have brought about a restoration of the saiga population. 
According to Adol'f (1950, 1952) a total of less than 3,500 saigas 
inhabited the interfluve of the lower Kuma and Volga in 1939, whereas in 
1950 herds of more than 1,500 specimens each were encountered; the total 
number of these animals living on the right bank of the Volga was several 
tens of thousands at this time. 

After World War II saiga again became common on the Terek-Kuma 
steppes (Babenyshev, 1948). 

In the severe winter of 1948-49 thousands of saiga died on the Astrakhan 
and Ciscaucasian steppes from cold and hunger. In the spring of 1950 
carcasses of saigas were found on the steppes of the eastern slopes of 
Yergeni in groups of five to eight (Kolesnikov, 1950). As snowfalls start 
large numbers of saiga rush to the south, toward the Terek and Sulak, 
and isolated groups reach the Dagestan foothills. Lavrovskii (1950) reported 
the gathering of tens of thousands of saiga north of the Chernye Zemli on 
tracts of fully-developed tall weeds. 

During blizzards, storms and deep cold in severe winters, saiga, like 
goitered gazelle, retreat into the tugai and reeds. Their appearance in 
the tugai along the Terek and Sulak is, therefore, not astonishing. When 
saiga undertake long migrations, going hundreds of kilometers northwards 
in search of moisture to quench their thirst during cold winters without 
snow, they are liable to die if they are suddenly caught by a snowstorm. 
As a result of such a migration, large numbers of saiga grazed in the corn 
and wheat fields of the Grozny Region in 1951. 

The present range and habits of saiga characterize it as a steppe animal 
which was driven into the desert by man. During the Pleistocene and 
Holocene it very probably lived in the forest-steppe and, in any case, 
360 inhabited thickets and reedlands of the floodplain, much like some African 

antelopes. Long migrations from 
forest to steppe and back, even 
in winters of unusually light snowfall, 
are confirmed by Litvin's cited 
report. In general, it may be said 
that the present stenotopic features 
of the species arc only superficial 
and, to a considerable degree, forced. 
The prospects for the continued 
existence of saiga in Ciscaucasia 
are poor; plowing and irrigation of 
vast steppe and desert areas will 
each year restrict further the areas 
suitable for the habitation of saiga. 

In 1950, 15 saigas were settled on 
Bulla Island, south of Baku: they 
could probably successfully inhabit 
any part of the semidesert of eastern 
Transcaucasia. 

Chamois — Rupicapra rupi- 
capra caucasica Lyd. Fossil 




FIGURE 144. Representation of a chamois on a 
silver bucket from burials in the Trialet ridge 
(Kuftin. 1941) 



356 



remains of chamois have frequently been found in Pleistocene and Holocene 
cave strata in Belgium, France, Germany and Italy, e.g., the Colomby 
grotto which is very rich in animal remains. The finds are almost always 
in mountain or piedmont areas. 

Only isolated occurrences of chamois have been found in Paleolithic 
cave settlements in western Transcaucasia, but the North Ossetian dzuars 
contain Middle Holocene remains. None has been found in either Pleistocene 
or Holocene strata in the vast piedmont plain region of the Caucasus, nor 
have they been encountered in Pleistocene deposits of Iran, Syria, Palestine 
and Lebanon. The contemporary distribution lies in the Pyrenees, the 
Cantabrian Mountains, the French, Swiss and Italian Alps, the Apennines, 
the Carpathians, the Balkan Mountains, Asia Minor and the Caucasus 
(Couturier, 1938). 

Thus both paleontological data and present zoogeography characterize 
the chamois as a mountain species occurring in the alpine fold systems 
lying to the north of the western Mediterranean and to the southeast of the 
eastern Mediterranean. 

Contour representations of chamois, depicting the animal with its chest 
pierced by arrows, appear on a silver bucket from Early Bronze Age burials 
on the Trialet (Figure 144). Since the chamois does not endure captivity 
well and could not have been transported far from its natural habitat by 
ancient tribes, this find indicates to archaeologists that the bucket was 
fabricated either in the Armenian Highland or in Asia Minor. The contemporary 
range of the chamois on the Caucasus is widespread (Map 88). On the 
Main Range, the distribution area can be traced from the west from the 
Goitkh Pass and Mount Semeshko (1,032 m) in the upper reaches of the 
Pshish (Dinnik, 1896, 1914a; Nasimovich, 1949b), with some occurrences 
observed (1955) in northwest Tuapse in the rocks of the Dzhubgi district, 
361 to the east on both sides of the Caucasus as far as Mount Baba-Dag in 
Azerbaidzhan. On the western Caucasus the chamois range is from 
250-300 m to 3,000 m above sea level. At the end of the last century and 
the beginning of the present one the species was very numerous in the 
reaches of the Belaya, Urushten and Malaya Laba rivers; as many as 
400 chamois could be observed daily and sometimes herds of 90 head were 
encountered (Dinnik, 1914a). Between 1926 and 1937, at least 4,000 chamois 
inhabited the Caucasian Reservation (Nasimovich, 1949b) and Zharkov 
reports as many as 12,000 (1940a). Winter concentrations are likely 
to occur on slopes with either a western or an eastern exposure. The animal 
is most frequently found in the middle altitudes of the subalpine zone — from 
1,500 to 1,700 m. 

In the central part of the northern Caucasian mountains, there are 
chamois habitations on the slopes of the Elburz and in pine forests of the 
ranges which border the upper Baksan, Chegem and Cherek rivers. The 
species is not found in the furthest chain of the Skalistyi, appearing only 
in Digoria and Ossetia. In Dagestan chamois live among rocks in an area 
of subalpine tall grasses and beech— hornbeam forests with a very humid 
environment on the ridges of the Chernye Gory to the middle course of the 
Argun. Their habitat in Urukh and the upper reaches of the Ardon is 
characterized by rocky areas grown with pine forests and stretches to 
Nogkau and Tseya. Chamois cannot be observed in central Dagestan which 
is deforested, and is found only in small numbers on the eastern slope of 
Mount Shakhdag. 



357 



On the southern slopes of the Main Range the species is common in 
Abkhazia and Svanetia, and it has found habitats on the very steep slopes 
of the upper Kodor, Ingur and Rion gorges in fir, spruce and maple forests, 
and particularly in the upper reaches of the Tskhenis-Tskhali, which are 
very rich in chamois. There are few to be found in South Ossetia or 
Kakhetia, but, further to the east, they are numerous in the beech and 
birch forests of the Katekh-Chai and Shin-Chai gorges and in the basin of 
the upper Turyan-Chai in Azerbaidzhan. 

Herds of 50-70 chamois can be observed in the Zakataly Reservation in 
August. In the 1930's the census of chamois in this reservation was 
approximately 1,000 head (Markov and Mlokosevich, 1935). The usual 
habitats of this animal are the edges of birch forests and very steep sod 
slopes grown with matgrass and fescue in those parts of the subalpine 
meadow-steppe with bedrock of black slate. On rocky slopes near Kurban- 
Efend in the Ismailly region chamois live even in the summer at altitudes 
from 800-900 m. In winter, the bulk of the population inhabits beech forests 
in the range of 800-1,200 m. 

The population and distribution of chamois is considerably less on the 
Lesser Caucasus. They are still found along the middle course of the 
Chorokh in the Artvin area and, further to the west, on the eastern Taurus, 
e.g., the Trebizond vicinity. There are no data on the habitat of chamois 
on the northern slopes of the Adzhar-Imeretia Range. At one time chamois 
was common in the beech forests and rocks of the Kura left bank in the 
Borzhomi area. In 1890 five chamois were penned in on a 130-ha-tract 
in the Borzhomi Reservation (Shil'der, 1892). On the Trialet ridge, the 
species can be observed in the BelyiKlyuch area, although it has been 
exterminated in other places. There were probably no chamois in the past 
and are none now on the deforested plateaus of Dzhavakhetia. 

As late as 1873, Bogdanov and still later Kalishevskii (1904) 
noted their presence on Alagez. Neither Burchak- Abramovich in 1944 nor 
Dal' in 1950 found any chamois there. In the last century the species 
inhabited the northeast margin of the Armenian Highland, and in the 1860's 
262 could be encountered along the Dzegam-Chai and Shamkhor-Chai rivers as 
far as the village of Barsun (Radde, 1899, p. 74). Folklorist Osipov (1898) 
mentions the presence of chamois in the mountainous areas around the 
village of Chaikend. By the 1930's it was no longer observable in Karabakh 
and on Shakhdag, although Sarkisov (l944g) maintains that it 
disappeared from the Murov-Dag only in 1930. Generally speaking, 
Sarkisov' s data on chamois distribution on the northern slopes of 
Shakhdag, on the Lori uplands and in northeastern Karabakh (I944g) require 
verification. 

The disappearance of chamois from Karabakh and Asia Minor cannot 
be explained only by human activity, because on the Murov-Dag and other 
ranges there are forests sufficiently deep for the survival of wild goat. 

There are no indications of chamois now or in the past in Talysh and 
on the Elburz, although they could certainly exist there. South of the 
Armenian Highland, chamois probably exists in relict forest areas. A horn 
of chamois from the mountains of the southern coast of Lake Van is 
preserved in the Zoological Institute of the Academy of Sciences. 



358 




FIGURE 145. Chamois lying on a scarp in the Katekh-Chai ravine 

Photograph by author, 1935 

Chamois is less well adapted to the intersecting relief of the mountains 
than is the goat. Its skeleton is lighter and its hooves are smaller. It 
avoids the rocky places frequented by Caucasian goats (although some 
naturalists think otherwise). Its principal enemies are big cats (panthers 
and lynx). It can survive in stable rocky areas which are 
inhabited by wolf and bear, because it is swifter of movement; it is even 
frequently found grazing side by side with bears. The seasonal and daily 
migrations of chamois are less pronounced than are those of goats. They 
do not migrate in winter to Dagestan where the snow cover is poor, and in 
summer they do not leave the forests for the glaciers, because they probably 
363 endure the attacks of bloodsucking diptera better than goats. Their 

attachment to biotopes of forest and mesophytic meadow is more stable than 
that of goats (Figure 145). On the whole, the chamois is a species adapted 
to the mesophytic areas of the mountains rising around the Mediterranean. 
It probably did not descend to the plain during the Quaternary because, 
being a specialized mountain species but without the speed of the roe deer 
or goitered gazelle, it could not move from range to range across the plain 
at that time without being destroyed by predators, particularly by wolves. 

The absence of chamois remains in plain burials and the data on its 
contemporary distribution and adaptations both indicate that the Caucasian 
and Alpine chamois derived independently from common ancestors which 
existed in the Tertiary on the folded uplands of the Sarmatian sea coasts. 

There is, thus, no room for discussion of a chamois migrationin the 
Quaternary to the European Alps from the mountains of Asia Minor and the 
Caucasus or, conversely, from Europe eastwards through the Balkans and 
Asia Minor. The Caucasian chamois should be regarded not as a subspecies 
of the alpine chamois but as an independent species. A comparison of the 



359 



two shows the Caucasian chamois to have distinctive morphological features, 
including a different voice — a long, hissing sound noted even by 
Dinnik (1896). 

The absence of chamois from Talysh and El'brus and the rapid decline 
of its population and shrinkage of its range in the 19th and 20th centuries 
on the Lesser Caucasus indicate that the contemporary ecological optimum 
of the species and probably the focus of its origin lie in the ranges 
surrounding the Black Sea to the south and east. 

The organization of a rational forest and hunting economy requires the 
renewed breeding of chamois in Karabakh — in the upper reaches of the 
Dzegam, in Shamkhor and in Talysh 

West Caucasian and east Caucasian goat — Capra caucasica Giild. 
ahd C. cylindricornis Blyth. Fossil remains of representatives of 
the subfamily Caprini have been found in Eurasia from the beginning of the 
Lower Pliocene, but on the Caucasus the only goat remains that have been 
found are from the Pleistocene and Holocene. The Pleistocene localities 
are all situated in western and central Transcaucasia on the margins of the 
contemporary range, often a distance of 50-60 km in a straight line from 
the present habitats of the species (Map 89). The explanation lies partly 
in the effects of the glaciations which forced the goats to lower altitudes, 
but even more in its widespread distribution in the Pleistocene (see 
Chapter II). 

To the south remains of local southern forms of С . sinaitica Erenb. 
have been found in the Paleolithic of Lebanon, Syria and Palestine and are 
recorded under the names of С . be den Wagn. , C. ibex L. and 
C. cf. nubiana Cuv. (Picard, 1937). Remains of Pleistocene goats are 
also reported from many mountain and foothill settlements in Western 
Europe — from Ireland and Belgium to Bulgaria. These remains, sometimes 
consisting of entire skulls, which are generally related to some species of 
the contemporary European goat — C. ibex L. , C. aegagrus L. , 
C. pyrenaica Schinz. — and to fossil species — I b e x priscus Wold., 
I. cebennarum Gerv. and others, have been found in Pleistocene as 
well as in Holocene strata (Woldrich, 1893; Wolf, 1938). A large goat, 
Capra sp. , inhabited the mountains of the Crimea in the Paleolithic 
(Gromova and Gromov, 1937). 
364 Carvings of Caucasian goats, especially of the east Caucasian type, 

were often produced in the Bronze Age; figures of goats appear on warders, 
pendants and seals (Figure 146). 

The contour representations on silver vessels from the Maikop burial 
show a similarity to west Caucasian goat (Figure 128). 

There are no paleontological or archaeological data on an earlier 
habitation of the Lesser Caucasus and the Taurus by Caucasian goat. 

Nevertheless, it seems very probable that such a habitation occurred, 
particularly if one compares Caucasian goat with species endemic to the 
area: Caucasian birch mouse. Promethean vole, pine vole, snow vole 
and the Caucasian black grouse. The feasibility of Caucasian goat existence 
on the Lesser Caucasus is demonstrated in the present day by the 
experience of the Borzhomi Reservation (Shil'der, 1892). 

The contemporary range of Caucasian goat includes almost all of the 
Greater Caucasus for 9 longitudinal degrees (between 39°45' and 
48°30'E). The chief habitat is on the slopes of the Main Range, but 



360 




FIGURE 146. Kobanian pole top 
of bronze representing east 
Caucasian goat — from North 
Ossetia (Uvarova, 1900) 



they are also encountered on pa.rallel ridges to 
the north and south and on spurs with southern 
exposures. 

In this vast area two contemporary species 
developed: the western — C. caucasica 
(C. caucasica severtzovi) and the eastern — 
C. cylindricornis, which differs from the 
western chiefly in the structure of the male horns 
(see Chapter IV). 

The distribution of the west Caucasian goat 
(Figure 147) now begins to the west of Mount 
Chugush (Nasimovich, 1949a), although in the 
1880's they could be found 30 km westward on the 
peaks of Fisht and Oshten (Dinnik, 1914a). In the 
Caucasian Reservation goats inhabit the 
Vodorazdel'nyi Range and its northern spurs. 
On the mountains of Bolshoi Pambak, Dzhuga, 
Akhtsarkhva and Yatyrgvarty, the habitats are 
somewhat separated in summer (Nasimovich, 
1949a). At the start of this century, goats were often encountered on the 
Skalistyi ridge and on Mount Achesbok. Temporary habitations by single 
specimens have been noted during the last fifty years on the summits of 
Dudugush, Chura and Khatsavita. 

The west Caucasian goat is also dispersed along the Vodorazdel'nyi 
Range and its northern and southern spurs from the upper reaches of Belaya 
and Mzymta to the upper reaches of the Balkar Cherek. Very few specimens 
are now found in the upper reaches of Teberda and Kodor, but in the 
Teberda Reservation, groups of 10-15 can be observed near the glaciers. 

Goats have completely disappeared from the upper reaches of the Ingur. 
To the east, they inhabit the southern slopes of Tetnuld in the upper reaches 
of the Mulkhra and single specimens are seen beyond the village of Ushkul'. 
The species is rare on the summits of Shoda and Laila, which are isolated 
from the Vodorazdel'nyi Range. There are numerous goats on the northern 
slopes of the southern ridges of El'brus. At the beginning of the 20th century, 

365 they lived on Mount Kinzhal, and at the beginning of the 19th century they 
were even found in the Pyatigor'e area on Mount Beshtau (Vietinghoff, 
1812). Beyond the villages of Tegenekli and Verkhnii Baksan groups of 
20-25 goats were observed in the 1950's. 

The species is also common in the Bezingi Cherek gorge, where in the 
1940's during the fall shepherds killed as many as 40 head. And it is even 
more common in the upper reaches of the Balkar Cherek on the slopes of 
Dykh-Tau. There in the course of a day in August 1948 the author 
encountered three females with young and two young males in the Dykhsu 
gorge, and observed numerous tracks and signs of goat-grazing near the 
glaciers. 

It is very rare in the upper reaches of the Urukh. 

366 The present distribution of east Caucasian goat lies to the east of limits 
drawn by the Ingur and Tskhenis-Tskhali headwaters on the southern slopes 
and the Baksan and Malka headwaters on the northern slopes. The skull of 
a female specimen from Teberda is preserved in the Zoological Institute 

of the Academy of Sciences. The Georgian Museum contains in its 



361 



collections a skull of a male from Teberda with horns of a type showing a 
transition from C. cylindricornis to C. caucasica, and a skull 
of an 11 -year-old male from Baksan showing a transition from 
C. caucasica to C. cylindricornis. Thirteen skulls taken from 
the northern spurs of El 'brus and Mount Kinzhal and now in the Zoological 
Institute collection also belong to the east Caucasian form. V. G. Khachvani, 
a Svan ofUshkul', near the Ingur source, owns a stuffed specimen, male, 
52 years old and typical of the east Caucasian form, which was killed near 
the summit of Shkhara in December 1947. From the same village Radde, 
in the last century, brought the horns of an aged goat of a transitional 
(hybrid '^ ) type between west and east Caucasian. 



(365) 




FIGURE 147. West Caucasian goats in winter 

Watercolor by E. Ya. Zakharov 



362 



In 1949 Nasimovich (1950) discovered in the Psygansu ravine five skulls 
of east Caucasian goat and one skull of west Caucasian goat which had been 
discarded by hunters. In finds in the upper reaches of the Bezingi Cherek, 
four skulls had horns of the west and central Caucasian type, and one had 
horns of the east Caucasian type. Two skulls found in the Balkar Cherek 
region both belonged to the eastern type. Of Dinnik's ten finds of skulls 
in 1887 in the Cherkess village of Kundyum two proved to be of the western 
form and eight of the eastern (Dinnik, 1890b). All 46 skulls taken from 
the Urukh and Ursdon gorges in the villages of Styr-Digor and Nogkau, 
which were examined by the author, belonged to the east Caucasian species. 

The Rekom dzuar in the Tseya gorge of the Ardon Basin contained 
298 pairs of horns, of which 292 belonged to the eastern type and only six 
to the western. 

From this location the range of the east Caucasian goat is continuous 
along the Vodorazdel'nyi Range (Map 89). 

According to data collected by the author and V.I. Naniev in North 
Ossetia in 1948, the east Caucasian goat can be observed on all the northern 
slopes of the Bokovoi and Main ranges and occasionally in the upper reaches 
of the Urukh, Ardon, Fiagdon and Gizel'don rivers. Herds of up to 100 head 
are encountered in the upper reaches of the ArkhonRiver and occasionally 
in the neighborhood of the villages of Dunta, Arkhon and Aksai. 

The species is now rare in the Grozny region and in central Dagestan. 
It was probably rare on the Gunib and Khunzakh plateaus even at the time 
of the Caucasian wars in the mid-19th century. Although now completely 
extinct in the middle reaches of Kara-Koisu, it undoubtedly could be found 
everywhere in that region in the last century. East Caucasian goat is 
rarely seen in the Samur and Lakskii areas, but is common in the Avar 
Koisu Basin on the Bogos Range, from which Geptner and Formosov (1941) 
report herds of 20 head in the 1920's. It is also numerous in the rocky 
upper reaches of Kara-Chai and Kusar-Chai on Shalbu-Dag and Shakh-Dag. 

There are now few goats on the southern slopes in the upper reaches of 
the Aragva and Alazan, and they are completely absent from the Telavi 
Range, even though they inhabited this area until quite recently. In the 
1930's some 200-300 head were counted in the Lagodekhi Reservation 
(Markov, 1940). Following some epizootic outbreak, their numbers were 
greatly reduced along the Belokan-Chai and Mazym-Chai (Vereshchagin, 
1938a). 

The Katekh-Chai ravine is the habitat, both in summer and in winter, 
for a particularly heavy population; herds of 200 and more head can be 
observed there. The census in the 1930's in the Zakataly Reservation was 
3,500-4,000 head (Markov and Mlokosevich, 1935; Vereshchagin, 1938a). 
The species is frequently found in cirques in the Shin-Chai and Demir- 
Aparan-Chai gorges, and is especially numerous on the southern border 
of Dagestan in the upper reaches of the Samur (Figure 148) and in the 
vicinity of the three high peaks of Bazar-Dyuzi, Baba-Dag and Shakh-Dag — 
^^'^ a fact noted even in 1896 by Satunin (Radde, 1899). In the 1930's the author 
observed there many herds of 30-40 goats, and occasionally aggregations 
of 250 or more. The extreme eastern margin of the east Caucasian goat 
range passes near Mount Aivasel 25 km east of Mount Baba-Dag. 
Today groups of 15-20 goats can be observed near Konakhent in the upper 
reaches of the Gil'gin-Chai. 



363 




FIGURE 148. Khalakhi Lake in the upper reaches of the Samur. 
The surrounding rocks are a favorite habitat of east Caucasian goat 



Photograph by author, 1937 



368 



The eastern Caucasian goat (Figure 149) ranges between wide altitudinal 
limits. This animal can be found in summer at altitudes up to 1,200-1,300 m 
in ravines of the Zakataly Reservation which are nearly inaccessible from 
the south. It frequents mineral springs and passes in valley bottoms at 
altitudes of 800-1,000 m. To the east above Vartashen and Ismailly its 
summer habitat is on the treeless slopes of the southern ranges at altitudes 
of 1,400-1,500 m. In winter the lower limit on snow-covered slopes drops 
200-250 m, but remains the same as in summer in the valley bottoms. After 
slaking their thirst at the mineral springs, the adult goats either stay near 
the riverbed at altitudes of 1,400-1,500 m or climb up to altitudes of 
2,800-3,000 m where they rest near snowdrifts, thus completing a daily 
vertical migration of 1,500-2,000 m. In August the goats often use passes 
at altitudes of 3,400-3,500 m. 

On the southern slopes the goat population of the upper reaches of the 
Alazan to the upper reaches of the Pirsagat divides in summer, because 
the entire alpine meadow belt is occupied by cattle herders with herds of 
sheep and cows guarded by dogs. One group of goats spends the entire 
summer in the forest, resting on taluses and rock overhangs; the other is 
to be found in the passes above the meadow zone on rocks and taluses. 

The west Caucasian goat of the Recent is now a mesophilous species, 
but its adaptation to the humid forest and to alpine meadows is probably 
a later, secondary phenomenon. 

The contemporary population of Caucasian goat is greater in the eastern, 
drier parts of the Caucasus; its preferred pasturage is in meadow-steppe 
rather than rich alpine meadows. It is characteristic of the goats of the 



364 



forest populations to take long rests on rocky scarps, taluses and in sunny 
forest clearings — a habit which may revert to the ancient ways of their 
ancestors whose mountainous habitats were treeless and more arid. 

The pattern of Caucasian goat behavior as it relates to firn-snow and 
glaciers is more clearly seen in summer than in winter. The animals 
generally frequent the firn in August and rest in the sun enjoying the coolness 
and the absence of flies (Vereshchagin, 1949a). They avoid frozen slopes 
and glaciers, and will look for footing on uncovered stones after a fresh 
snowfall. It is rare that they will attempt a crossing over a large glacier 
area. During the author's crossing of the huge glaciers near Mount 
Laborda east of Mount Shoda in August 1948 no goat tracks were observed 
on the glaciers. According to observations of hunters (M. G. Ivkin and 
others), large herds of goats migrate each year in November-December 
from Lagodekhi and Belokany to Dagestan, always using the snow -free 
crests of ridges . 

The general outline of the Caucasian goat ranges has changed little during 
the last 150 years. Presumably during the Caucasian wars in the mid- 19th 
century there were more goats killed than there are now. The population 
declined with the establishment of Russian settlements in the Ciscaucasian 
piedmonts and the introduction into the mountains of the four-barrelled 
Berdan rifle and the triple -barrelled rifle at the start of this century. 

The decline was progressive during the first half of the 20th century 
throughout most of the goat ranges. One exception was the hunting preserve 
of the Grand Duke in the Kuban Basin where the animals were well-guarded. 
The protected environment of the Caucasian and Zakataly reservations 
brought about an increase in the goat population in these areas beginning 
with the 1930's. 

Goats fled from harassment in Fisht and Oshten in the 1880's, and 
disappeared from Mount Achishko near Krasnaya Polyana in 1908 
(Nasimovich, 1949a). Villagers of Ushkul' killed 200-250 goats in the area 
of the Ingur sources over a 35-40-year period in the last century. Their 
sons, using improved weapons, killed no more than 15-20 in the same time 
span in the first half of this century. 

By the beginning of the 20th century, goats had been routed from the 
furthest ridges of the Skalistyi in the central part of the northern Caucasus, 
particularly from Kinzhal and Barmamyt (Dinnik, 1914a). Dronov collected 
thirteen skulls in the upper reaches of the Malka as late as the 1880's. 
They have probably been extinct inPyatigor'e in the Beshtau region since 
the beginning of the last century. 

Judging fronn skulls found in the dzuars of Lesgor and Digorized, the 
east Caucasian goat inhabited Digoria early in this century on the Skalistyi 
ridge which cuts off the longitudinal valley of the Doniserdon from the 
north. Since no fresh skulls have been found in these sanctuaries, the 
species must have disappeared by the 1930's- 1940's. 

The approxinaate annual yield of Caucasian goat in the 1930's was 
probably no less than 4,000 head (Figure 150). 
369 Caucasian goats, then, are not to be considered as on their way to 

extinction, even though there is no control over hunting in the mountains. 
The existence of three reservations — Caucasian, Teberda and 
Zakataly- Lagodekhi — is a guarantee of a continuing, thriving population. 



365 



370 



The development of cattle breeding, accompanied by the extermination 
of wolves and panthers, had both a positive and a negative effect on the 
Caucasian goat population. In the early stages of cattle herding, wolves 
preyed on the domestic cattle in the mountains in the summer and followed 
the herds to the plains in winter, thus diminishing the danger to the goat 
population. However, as cattle breeding was further developed and pastures 
were enlarged, cattle diseases such as scab, foot-and-mouth disease 
and others spread to the goat population (Dinnik, 1914a; Vereshchagin, 
1938a; Nasimovich, 1941). 




FIGURE 149. East Caucasian goats in summer 

Watercolor by E.Ya. Zakharov 



366 



The facility with which Caucasian goats can be crossed with domestic 
species, domesticated and acclimatized to the plains, leads to an anticipation 
of further development of both species through controlled expansion. They 
will probably be bred in the Karabakh uplands, on the Shakhdag, and on the 
Trialet and the Adzhar-Imeretia ridges, as well as in Talysh and El'brus. 




FIGURE 150. Horns of Caucasian goats mounted on a veranda of an Ossetian house in the Urukh gorge 

Photograph by author, 1947 



371 



The economic effects to be obtained froi-n the hybridization of Caucasian 
goats with domestic goats could be considerable (Vereshchagin, 1938a). 

Taking into account the endemicity and morphological isolation of 
Caucasian goats from other representatives of the genus (see Chapter IV), 
the species should be regarded as local in origin, springing from Caucasian 
Tertiary ancestors. 

Bezoar goat — С ap r a aegagrus Erxl. Fossil remains of this 
species are recorded for Pleistocene and Holocene cave strata in Italy, 
Lebanon, Syria and Palestine (Picard, 1937; Wolf, 1938). They have been 
found in the Caucasus only in postglacial strata in caves in the Araks gorge 
(Polyakov, 1882) and on the Saraibulakh ridge, in Bronze Age strata of the 
Teishebaini fortress ruins near Yerevan (Dal', 1940b, 1952), and in dzuars 
in Tushetia and Khevsuretia. 

Bronze Age drawings on limestone rocks in southeast Kabristan west 
[? south] of Bakushow a similarity to Bezoar goat (Figure 151), and may 
indicate a recent habitat for this species on the lower ridges of Kabristan 
near the Caspian Sea. The collections of the historical museums of Moscow, 
Makhachkala and Grozny contain exquisite bronze figurines of Bezoar goat 



367 



from Dagestan (Figure 152) which are associated with the cultures of the 
Kobanian and Kayakent-Khoro-Chai period (first half of the 1st century В . С. ). 




372 



FIGURE 151. Representations of goats on rocks of Beyuk-Dash south of Baku 

Photograph by author, 1945 

The contemporary distribution of Bezoar goat extends beyond the 
Caucasus to the mountains of southwest and central Asia, from Crete to 
western India through Turkey, Iran and Afghanistan. There are three 
isolated ranges on the Caucasian Isthmus: the eastern part of the Greater 
Caucasus and the central and the eastern parts of the Lesser Caucasus. 
The upper reaches of the Alazan, the Argun and the Gerdyman-Chai, which 
is to say the arid, treeless part of the country, form the limits of the range 
on the Greater Caucasus (Map 90). In the 1920's the species inhabited the 
Lakskii area of Dagestan, and, according to Dinnik (1914a), could be 
encountered 30 km from Gunib at the start of the century. 

The Bezoar goat of Dagestan is rarely seen in the forested, humid 
slopes of the Main Range between Lagodekhi and Shemakha and in the passes 
of the divide — perhaps once in two or three decades. 

It is common in the upper Avar Koisu area, in the vicinity of the villages 
of Tlyadal, Kidero, Chorod and Takhota, where it inhabits pine and beech 
forests. Along the upper Terek and Aragva it is not now observable and, 
to judge from the complece absence of skulls of the species in Ossetian 
dzuars, it was probably never an inhabitant of this area. 

The range on the Lesser Caucasus is more extensive: it includes on the 
southwest the fir forests of the upper Chorokh, certain parts of the 



368 




FIGURE 152. Bronze figurine of Bezoar goat 
(Makhachkala Museum) 



373 



Adzhar-Imeretia Range, the western spurs of the Trialet in the Borzhomi 
District and the southeastern part of the Armenian Highland. The species 
is common on the Shakhdag and in the Kedabek and Kalakent areas, e.g., 

in the upper reaches of the Shamkhor 
near the Kashkar summit (3,379 m), 
on Mount Kyapyaz, beyond Lake 
Gek-Gel and below the peak of the 
Murov-Dag (3,740 m). 

In 1940 a small number of these 
goats inhabited the taluses in beech 
forests on the southeastern slope of 
Mount Kechel-Dag at the eastern 
extremity of the Shakhdag ridge. 
The guard at the sawmill there killed 
seven or eight goats each year. 
Mount Kirs (2,743 m) above 
Stepanakert was the easternmost summit 
I inhabited by Bezoar goats in the 
1930's. 

The habitat of this animal in the 
central and southern Armenian Highland 
lies in rocky mountains from the Azat 
River valley to the Zangezur, Megri and Bargushet ridges. North of the 
upland the range is limited by the southern slopes of the Gegam and 
Vardeniss ridges (Dal', 1951b). The Bezoar goat is frequently observed 
on the Urts and Aiotzdzor [Daralagez] ridges and on the rocky massifs 
close by. In 1950 Dal' (l951b) counted 124 goats in five aggregations on 
33.5 km^ on the Urts ridge. 

In the 1870's goats could still be encountered near Elenovka on the Sevan; 
Polyakov collected one skull from this region. 

In the Nakhichevan A.S.S.R. in 1947 the summits of Alindzha-Dag and 
Ilyanlu-Dag in the vicinity of the villages of Abrakunis, Boyanur and 
Khanagya provided a habitat isolated by the upland semidesert. During the 
historical era, goats were repeatedly routed from these heights by ancient 
Armenian tribes which used the summits as natural fortresses. Bezoar 
goats are more adaptable than the Caucasian type; they often cross vast 
desert areas easily and, therefore, have a great migratory potential. 

On the Zangezur ridge and its southern spurs the goat population spreads 
eastwards as far as the Araks gorge. It is particularly numerous in the 
Negram Mountains between Dzhulfa and Ordubad at altitudes of 
1,000-1,200 m. In the Araks gorge it is distributed on the southeastern 
slopes of the Megri ridge as far as Okhchi-Chai. In 1947 a small goat 
population inhabited desert rocks near Akluis and Ordubad. 

There are references in the 14th-century writings of Rashid ad -Din to 
the Bezoar goat as an inhabitant of Talysh, where it could still be 
encountered in the high mountains, e.g., on the Kyz-Yurdy summit, 
(Radde, 1899, p. 75) in the mid-l9th century; later it was no longer 
observed there. 

To the south the range of this goat extends into the dry uplands of Asia 
Minor and Iran. Collection specimens were taken from Erzurum, Van, 
Kars, Ararat, Tabriz, Sabalan and Damavand at the end of the last century 



369 



and the beginning of the present one. In the U.S.S. R. , the Bezoar goat 
also inhabits the Kopet-Dagh and the Bolshie Balkhany. According to 
Sarkisov (I944e) it can be observed in Iran throughout the mountains of 
Azerbaidzhan, Kurdistan, Kermanshah and Luristan, and to the south in 
the provinces of Yazd and Shiraz, in Makran, Baluchistan and as far as 
the Sulaiman Range. 

The most usual habitat of the species is an arid, rocky mountain biotope 
with buckthorn, hawthorn, juniper and other xerophytes. The vertical 
distribution of the contemporary Bezoar goat on the Caucasian is from 
900 to 3,500 m, but it manages very well at altitudes lower than that, even 
down to sea level. 

The adaptation of the present range to the eastern Mediterranean area 
and the life pattern of the Bezoar goat indicate a protracted evolution 
of the species in a dry, hot rocky mountain environment at relatively low 
altitudes. The penetration of this goat into the Greater Caucasus probably 
proceeded northwards from southwest Asia in one of the xerothermic 
periods of the Pliocene or Pleistocene, similarly to that of hamsters of the 
genus Mesocricetus. The Surami Range and the steppe -like piedmont 
of eastern Ciscaucasia would have provided a likely route. The subsequent 
isolation in the arid eastern part of the Greater Caucasus can be explained 
by the glaciation of the ranges and the development of a wide mesophytic 
zone of forest and meadows to the west. The occurrence of the species in 
Dagestan at a later time, e.g., Lower Holocene, is less probable. 

Anthropogenic influences on this goat only came into play in the last 
decades with the development of cattle herding and the introduction of 
long-range guns into the mountains. The decrease in the range is 
particularly apparent today on the southeastern Armenian Highland. 

In an ideal hunting economy on the Caucasus, the species would have 
an unlimited future. Its range could be extended by controlled settlement 
along the longitudinal valleys of the central Caucasus, e.g., in the Malka, 
Baksan, Cherek, Urukh and Fiagdon ravines where there are places on the 
cuesta scarps of the Skalistyi which provide favorable conditions for its 
habitation. 

Sheep: Argali — О vis cf. ammon L. — and Armenian mouflon — 
О vis gmelini Blyth. Argali remains are reported from Western Europe 
for the Pliocene to the Upper Pleistocene, and mouflon remains for the 
Pleistocene to the present. 

On the Caucasus occurrences of large sheep of the argali type (Figure 153] 
are very rare and have only been found in Middle and Upper Pleistocene 
strata of the Transcaucasian mountains and plains. Finds of mouflon 
remains are even more rare and have been made only in the Paleolithic 
of the Lesser Caucasus and the Mesolithic of Dagestan. 

The localities closest to the Caucasus from which finds of Pleistocene 
argali have been recorded are caves in the Crimea (Acheulean and 
Aurignacian strata). Bones of mouflon-type sheep, O. cf. argaloides 
Nehr. , were found in Acheulean-Mousterian strata of the Kiik-Koba cave 
and in Aurignacian strata of the Adzhi-Koba cave (Gromova, 1935a; 
Gromova and Gromov, 1937). In spite of the suitability of the topography 
of Palestine, Lebanon and Syria, remains of sheep from Paleolithic and 
Neolithic cave strata in these regions are unknown (Picard, 1937). 



370 




FIGURE 153. Remains of argali 

1,3 — metacarpal and humeral bones from Middle Pleistocene asphalts near Baku; 2 — M3 
from Upper Pleistocene diluvium near Maragheh 

A tooth of a large argali was found in Quaternary loams of northwest 
Iran near Maragheh, and remains of О vis sp. were found in caves in the 
vicinity of Urmia and Asterabad (Coon, 1951). 

There are no occurrences of argali remains in Holocene strata of the 
Caucasus. 

Aside from the Mesolithic of Dagestan, bones of a small post-Pleistocene 
sheep, O. cf. gmelini, have been found only within the linnits of the 
contemporary range of this species: in the ruins of the Karmir-Blur 
fortress near Yerevan (7th-5th centuries B.C.) and in cave strata of the 
Saraibulakh ridge (Dal', 1941, 1952) (Map 91). 

Depictions of sheep are scant in the ancient literature and representative 
art of the Caucasian peoples. The earliest are to be found in sculptures. 
A mouflon head carved on a belt buckle from the Trialet burials in southern 
Georgia has been related to the second millennium B. C. (Kuftin, 1941). 
Bronze pendants in the form of sheep with argali-like horns are quite 
common in Kobanian burial and settlement sites. These representations 
give an impression, however, of domestic rather than wild animals, as 
they do not have the force or vivid quality of untamed animals (Figure 154). 
Argali probably disappeared from the Caucasus in the Holocene, not even 
surviving in central Dagestan where environmental conditions were most 
favorable . 

Gold articles have been found in Scythian burials of western Ciscaucasia, 
Kellermess and other localities which bear traceries and relief 
representations of small-horned sheep of the mouflon type. Similar 
representations on so-called "Sassanid metal" from Transcaucasia date 
from 220-657 A.D. Silver plates in the Hermitage collection show a mounted 
hunt of Chosroes I and Shapur II in pursuit of fleeing sheep (Orbeli and 
375 Trever, 1935). The area in which the plates were produced —the Iranian- 
Anatolian Plateau — corresponds to the contemporary rage of southwest 
Asian sheep. 



371 



The present range limit of mouflon-like sheep in the U.S.S.R. is reached 
in southern Transcaucasia (Map 91). 

Horns and pelts of this species collected from Yerevan, Nakhichevan, 
Negram, Dzhulfa, Darry-Dag and the village of Puzian in Zangezur are 
preserved in the Georgian Museum. The nearest Iran localities to produce 
specimens are Kotur-Dagh, the vicinity of Khoi and the island of Koyun-Dagh 
on Lake Urmia. 

The collections of the Zoological Institute of the Academy of Sciences 
contain horns found by Academician Brosse in Alagez in 1849 and a skull 
from the same mountain near the village of Mastart found by Pfizenmair 
in 1911. There are also skulls and pelts collected throughout Armenia; 
from the Nakhichevan A. S. S. R. (the environs of the village of Shakhbuz, 
Dzhamaldin, Arazia), and from Turkey (Mount Ararat, the environs of 
Bayazid, the village of Gurdzhi-Bulakh, Pir-Reshid Mountain 60-80 km 
from Van Lake). The collections of Mil'kovich (1910), Brandt (1879). 
von Vik (1914), Khanykov (1851) and Vvedenskii (1908) contain horns, skulls 
and pelts from Kara-Dag to the south of Ordubad, the Zorskie Mountains, 
Lake Urmia, Mount Sabalan and the environs of Tabriz. 

The distribution of sheep in southern Transcaucasia has diminished 
rapidly since the last century. Chopin (1852) comments on the presence 
of sheep in Alagaz, but contemporary authors — Sarkisov (1941, 1944a, 
1944b) and Dal' (1945, 1949b) — note that the distribution area ends east 
of the Zanga River on the Saraibulakh and Daralagez ridges. 

In 1947 the species was rare in 
the environs of Abrakunis, Dzhulfa, 
Negram and Ordubad and on the Darry- 
Dag and Ilyanlu-Dag mountains, and 
only a small number inhabited the 
Zangezur ridge, particularly the 
vicinity of the villages of Bichenakh 
and Shakhbuz. 

The contemporary range of 
mouflon in Transcaucasia is 
associated with the upland steppes 
where groupings of dense xerophytic 
grasses and sibbaldia predominate, 
with formations on mountain slopes 
of thorny astragali in the middle 
zone, and with salt— wormwood 
groupings in the Araks valley. The 
summer habitat of the species is 
in the high mountains and their 
descents into the warmer valleys 
are made mainly in winter. 
It can be supposed from the 
general distribution of sheep in Eurasia (Nasonov, 1923) that argali and 
mouflon are newcomers to the Caucasian Isthmus. Their range expanded 
from a southern focus after the development of a xerophytic landscape. 

The distinctiveness of argali-like sheep and mouflon-like sheep cannot 
be considered as mere geographic variations of a single species, since the 
divergence between the two can be traced as far as the Middle Pleistocene 
(Vereshchagin, 1953c). 




FIGURE 154. Bronze heads of argali-like sheep 
from Kobanian burials of North Ossetia 
Uvarova, 1900) 



372 



The immediate future of wild sheep on the Caucasus, like that of other 
ungulates, will depend mainly upon the effectiveness of hunting regulations. 
376 The argali-like sheep should be settled in central Dagestan in Gunib, 
Botlikh and Khunzakh. 

Primitive and Caucasian bison — В i s о n sp., B. cf. schoetensacki 
Freud., B. priscus Boj. , B. bonasus caucasicus Sat. Bison 
remains occur in Eurasia from the Upper Pliocene, the earliest finds — 
horn stems of the small bison Bison sp., B.cf. schoetensacki— being 
in Upper Pliocene Apsheron clays in Kabristan (eastern Transcaucasia) 
and in conglomerates on the Taman Peninsula (Burchak- Abramovich, 1949; 
Vereshchagin, 1957a). 




FIGURE 155.' Caucasian bison 



Middle Pleistocene remains of a larger bison — B. priscus 
longicornis — and Upper Pleistocene remains of a smaller one — 
B. priscus deminutus — have been found in many localities in 
Ciscaucasia and western Transcaucasia. Most of the finds were made in 
lacustrine -fluvial and cave strata (Map 92). 

The form was rare in the Pleistocene of eastern Transcaucasia and of 
the Armenian and Iranian uplands, even being replaced by primitive tur. 
Its absence from the Binagady complex is especially puzzling. To the 
southwest, however, it occurs quite commonly in caves of Lebanon, Syria 
and Palestine in strata from the Acheulean to the Neolithic (Picard, 1937). 

The trend toward a smaller size in the bison of Eurasia which can be 
observed at the end of the Pleistocene was reinforced and fixed by the 
harassment of man and the aridity of the landscape. There is no reason to 
attribute this phenomenon to "deterioration" of the climate in the Upper 



373 



Pleistocene, as zoologists and paleontologists have often done. The cooling 
of the climate and the development of meadow formations caused an increase 
in the size of bison in the Middle Pleistocene. The contemporary Caucasian 
377 bison (Figure 155) is undoubtedly a direct descendant of the local Pliocene- 
Pleistocene bison. 

At the same time that the bison size diminished in the Upper Pleistocene 
and postglacial periods, the populations became isolated in river valleys, 
on high plateaus and in broadleaf forests on the mountain slopes . The Caucasian 
bison has not retained the steppe habits of its Pleistocene ancestors and 
American relatives. It has become an inhabitant of the mesophytic forest- 
steppe and the mountain forest zone. Its remains have been observed in 
certain localities of western Transcaucasia and the Lesser Caucasus. 

In the Middle Ages the Caucasian bison population was probably linked 
to the bison population of Eastern Europe by way of the wooded valley of 
the Don. 

The different fates of the Eurasian bison and the American bison in the 
Holocene were caused by different anthropogenic effects (Vereshchagin, 
1956). The survival by large bison and turs of the destructive mounted 
assaults of hordes of Khazars, Polovtsians, Kipchaks and Mongols would 
have been strange indeed. They did survive, however, into the 16th century 
in Great Lithuania and Mazovia (Vizhener, 1890 edit. ) where they were 
hunted with spears ("metabulo"), and into the 18th century in western Siberia 
(Bell, 1776). Pidoplichko's assertion (1951) that the European bison of the 
Pleistocene was a permanent inhabitant of the steppe is not well founded. 

It is strange that among ancient Caucasian artifacts scuptural and graphic 
representations of bison are almost unknown; only some unrealistic 
figurines of bulls that somewhat resemble bison have been found among 
bronze trinkets of the Kobanians. References to Caucasian bison in literature 
date from the Middle Ages. 

Rashid ad-Din (1946 edition) writes that when Abaga-Khan was wintering 
in Arran in 1275-1276, he and five farsang (Parsees) from Shahrud hunted 
the "mountain buffalo" in the forest. In describing a hunt of the Ghazan- 
Khan in Talysh in 1301-1302 (pp. 188-189), this chronicler reported: "After 
this the warriors organized a round-up and chased the game — mountain 
buffalo, dzhurs, wild goats and asses, jackals, foxes, wolves and bears 
and all kinds of wild animals and beasts of prey — inside the fence until 
they all were gathered in that enclosure. " The author probably referred to 
roe deer ("dzhuyurs") as "dzhurs, " and to Bezoar goats or goitered 
gazelles as "wild goats. " 

It is noteworthy that tales of wild bulls in Talysh and the El'brus 
Mountains were repeated up to the 19th century (Menetries, 1832). Brandt's 
critical survey (1867) added nothing new on this subject. It is possible that 
the extinction of bison in the forests of El'brus and Talysh occurred at 
approximately the same time as it did in the forests of the central 
Caucasus, i.e., in the 18th- 19th centuries. 

The earliest literary reference to bison in Circassia is probably that 
of Jean de Luc (1625, 1879 edition) and Arcangelo Lamberti's (1954) 
is the first allusion to the species in Abkhazia. 

It was known in Moscow that the bison existed in Kabarda, Ossetia and 
Ingushetia long before the publications of Academicians Lovich, Giildenstaedt 
and others who are considered the earliest discoverers of bison on the 
Caucasus (Bashkirov, 1940). 



374 



Page 7994 of the 10th volume of "The Complete Collection of Statutes of 
the Russian Empire, 1649" gives this directive: 

"Written Order from the Cabinet of Her Majesty the Empress Anne 
Holstein [daughter of Ivan V] to the Astrakhan Obercommandant — on the 
yearly catching of various living animals to be sent to the Court and to the 
Izmailovo menagerie. . . It is known to us that there are in Kabarda wild 
bulls and- kdosy [bison] which they call dombai, for the sake of which you 
3'78 shall do your best and not spare money from the Treasury in order that the 
local princes shall command the catching of five or ten such bulls and young 
cows and send them to the Kizlyar fortress. There they should be lured 
with bread, and when they are habituated to it, you shall send them to 
Moscow together with the other animals. Write this to the Gori Shah and to 
the commandant of the Kizlyar fortress and to the Elmurza of Cherkassia, 
asking them to do their utmost for the capture and dispatch of these bulls 
and cows . " 




FIGURE 156. Bison skulls in the Digorized cave 



Photograph by author, 1947 



It is possible that bison still inhabited North Ossetia at the time of 
Giildenstaedt's journeys in 1770-1774. 

Dinnik (1890a) was able to find old men of the village of Zadalesk who told 
him how their fathers and grandfathers killed the bison-dombais with iron 
bullets. There is an inscription carved on one of the skulls of the Digorized 
cave which reads "1833" and may signify the date when one of the last bison 
was killed (No. 10, North Ossetia Pedagogic Institute) (Figure 156). 



375 



By the middle of the 19th century bison survived only in the naountains 
and forests of the northeastern Caucasus, from which they gradually 
disappeared in several stages. The range limit in the north was probably 
the longitudinal course of the Kuban, including the mouth of the Malyi- 
Zelenchuk, and in the west the Pshish River. A comparative reading of 
recorded folklore led Bashkirov (1940) to assume that bison very possibly 

о '7 Q и i. J 

•-' ' ^ inhabited the upper reaches of the Shakhe, Sochi, Mzymta, Psou, Bzyba, 
Kodor and Ingur. 

In the 1870's there remained a total of more than 2,000 head of 
bison. 

Little by little, the Caucasian wars and the settlement of Cossacks in 
the foothills drove the bison to the upper Belaya-Laba interfluve. The 
decline in population and the shrinkage of the range continued without 
interruption as cattle breeding and hunting expanded and deforestation 
proceeded. 

By the 1890's the Caucasian bison population was estimated at 
500-700 head (Satunin, 1898), and by 1918 Kulagin's census was only 442. 




FIGURE 157. Bison (male) in a fir forest 

Photography by D. F. Filatov, 1909 

The establishment in the 1890's of the Grand Duke's hunting preserve 
in a 522-dessiatina tract [1 dessiatina — 2.7 acres] of the Kuban and the 
control of poaching in this area halted the decline to a certain extent. 

Filatov's investigations (1910, 1912) revealed a further contraction of 
the bison range; this map shows a distribution area only in the upper 
Belaya-Bolshava Laba interfluve. He placed the bison population at that 
time at several hundred, mostly inhabitants of fir forests (Figure 157). 



376 



After the 1917 Revolution, the Kuban hunting preserve was overrun 
by cattle herders, lumbermen, army deserters and hunters armed with 
triple -barrelled rifles who destroyed nearly all the Caucasian bison in 
that area. 
380 In 1919 an epizootic, probably endemic to domestic cattle grazing in the 
mountains, broke out among the bison and killed virtually all the rest. 

Only 50 animals survived in 1920, at least 20 of them in the Dishi- 
Khamyshki district, and these too were subsequently driven off by poachers. 

The establishment of the Caucasian Reservation in 1924 could not save 
the bison because of the number of small arms among the local population 
and the difficulties of protecting the animal in the mountains. 

Several bison killings are recorded for the years 1921-1926: near the 
Cossack village of Lineinaya, 50 km west of Maikop (1921), near Mount 
Gefo (1923), and in the upper reaches of Kodor on Alous and Mastakan 
(1925-1926). The latter report from Mount Alous describes the killing 
of three bison, probably the only survivors, by Imeretian herdsmen, and 
is the last one on record (Bashkirov, 1940). 

With the year 1940, the history of the hybrid Caucasian-American 
bison begins on the Caucasus (see Chapter VI). These bison possess a 
greater vital capacity and can live in the mountain forests throughout the 
northern Caucasus and Transcaucasia. 

Primitive bull* — В о s trochoceros Meyer, B. mastan-zadei 
Burtsch. , B. primigenius Boj. , B. minutus Malsb. It is customary 
for a history of the genus Bos to commence with the deposits in northern 
India in which many Pliocene and Pleistocene fossils of extinct species of 
Bovidae have been preserved. 

Like bison, the genus displayed on the Caucasus a successive series of 
related forms, which, however, were not as clearly defined as the bison 
species. In addition, some species represented lateral phylogenetic 
branches . 

The history of the primitive European-type Bos — a branch of 
B. primigenius — begins on the Caucasian Isthmus with the Middle 
Pleistocene. 

The most ancient find is the skull of a giant B. trochoceros in 
Middle Pleistocene sands of Adzhi-Eilas south of Yerevan (Avakyan, 1946). 
Later, Upper Pleistocene finds are recorded for the Apsheron Peninsula 
bitumens (Bogachev, 1925b; Burchak-Abramovich, 1951d, 1952d), for the 
Armenian Highland and for Ciscaucasia (Map 93). There the history of 
Bos on the Caucasus seems to be broken, although burials of the second 
millennium B.C. (Hittite ? culture) on the Trialet ridge near Tsalka and 
Kirovokan revealed skulls, metapodia and phalanges of bulls of a size close 
to the Holocene wild bull. 

Related forms on the Caucasus are those of the Middle Pleistocene 
B. mastan-zadei from the Apsheron bitumens (Burchak-Abramovich, 
1952d), and of the Holocene B. cf. minutus, a small postglacial bull 
from the sands of the Sevan coast (Dal', 1950a). These are isolated 
localities. The Caucasian B. mastan-zadei did not become an inhabitant 
of forests and did not migrate to the mountains — an interesting point of 

* [The Russian text uses "tur" here, but to avoid confusion between bull and goat (since, in English, "tur" 
can also mean Caucasian wild goat) we have substituted "primitive bull" in the translation as the common 
name for the genus Bos where it is apparent that the reference is to one of the primitive species of Bos.] 



377 



difference from the bison. The distinguishing ecological characteristics 
of these Quaternary species can be observed in the distribution of their 
remains in Eastern Europe and on the Caucasus. 

For instance, to the north of the Caucasian Isthmus, remains of primitive 
bull are well authenticated from alluvial and diluvial deposits of the Russian 
Plain river valleys. 

The Bos population was less than that of bison throughout its range. 
The lower percentage of remains of В о s is particularly apparent to the 
381 southeast. Table 74 shows the breakdown between authenticated bison 
and bull skulls collected mainly from river banks and now in museum 
collections. 

TABLE 74. Distribution of finds of bison and bull skulls on the Russian Plain (data from 
museum collections) 



Museums 


Bison 


Bull 


no. 


^ 


no. 


7o 




23 
13 
11 
36 
67 
38 
6 


85.1 
76.5 
84.5 
81.8 
95.7 
74.5 
85.7 


4 
4 
2 
8 
3 
13 
1 


14.9 


Ural 


23.5 




15.4 


Saratov 

Astrakhan 


18.2 

4.3 

25.5 




14.3 







This distribution confirms to some extent Pidoplichko's statement (1951) 
that primitive bulls inhabited more humid landscapes and biotopes than 
did bison. 

In Europe, the bull was mainly an animal of the mesophytic forest 
particularly toward the end of its existence. On the Caucasus and in 
southwest Asia this was not the case: it inhabited humid valleys and 
meadows of open plateaus and, of course, was easily exterminated. Recent 
experience with cattle breeding in the plains of central Asia shows a 
parallel: the spring ephemeral growth of the desert and semidesert suffices 
for pasturage for only a short time; later, from June onward, grazing is 
concentrated in meadow -swamp formations on the shores of lakes that are 
not stream-fed (e.g., the Chushka-Kul lakes near the city of Turkestan) 
and in the river valleys. Similar feeding habits can be observed in the 
coarse -grass steppe: as the dense, scrubby grasses become coarser, the 
cattle migrate from the steppe to the ravines, gulleys and humid floodplains. 

Attracted as the genus is to meadow formations, the primitive bull 
probably found its optimum biotope in the southern regions — southwest 
Asia and the Caucasus — at the time of cooling and humidification of the 
climate. 

The habitation of the Apsheron Bos on the Pleistocene semidesert — 
the eastern Transcaucasian steppes — was probably seasonal, that is, 
confined to the period of winter-spring ephemeral growth and not including 
the summer. Primitive bulls did not inhabit the forested and swampy 



378 



Transcaucasus, but, to the south, they were common on the wooded and 
steppe -like plateaus of southwest Asia. 

Remiains of Bos primigenius found in Mesolithic strata of the Belt 
cave near Asterabad Bay should probably be correlated with descendants 
ofB. mastan-zadei, a more xerophilous form than European Bos. 
(The same correlation holds for goitered gazelle remains —Coon, 1951.) 
B. nomadicus Falc. , remains of which are recorded for Lower Neolithic 
strata from the site of the ancient town of Annau near Ashkhabad (Duerst, 
1908), was probably a descendant of the Middle Pleistocene Apsheron Bos. 

Two forms of Bos, one large and one small, probably existed on the 
Armenian Highland at the end of the Pleistocene. Judging by the finds of 
382 bones of bulls in the Neolithic strata of Urartu, it can be assumed that both 
forms were domesticated. 

There is no documentation of the existence of В о s on the Caucasus 
in our time. 

Remains of В о s have been found, extraordinarily enough, in kitchen 
middens of ancient towns on the Russian Plain. However, to interpret this 
find to mean that the wild Bos was a dangerous enemy of prehistoric man 
(Gromova, 1931, p. 362) and that man began "to avoid the hunting of this 
dangerous prey" when domesticated animals were available is very naive. 
We have Shumerian and Assyrian bas-reliefs showing scenes of hunting to 
contradict this interpretation. 

The explanation lies rather in the sharp decline in the population of this 
animal in the Neolithic and in the habits of the European Bos which made 
it difficult to hunt. The bison which lived in the forest and was no less 
dangerous was hunted throughout the Holocene, In the Middle Ages Bos 
were encountered only in isolated localities and were often specially 
protected (Vizhener, 1890 edition). 

The Bronze Age and, particularly, the Scythian epoch on the Caucasus 
have furnished us with representations of bulls: a silver vessel from the 
Maikop burial (second millennium B.C. ) bears a contour drawing — two 
bronze figures are designed in the form of bulls, and flat gold plates show 
a bull in profile (Figure 158). 

It is likely that contour drawings of bulls and cows on rocks (Figure 159) 
in Kabristan south of Baku, probably belonging to Roman time, depict 

domesticated animals (Vereshchagin and 
Burchak-Abramovich, 1948). A bronze figure 
of a powerful bull, similar to the primitive 
species, was found in a Kobanian burial in 
Ossetia. In the collection of the Historical 
Museum of the Academy of Sciences of the 
Azerbaidzhan S.S.R. are many signets and 
rings with relief representations of bulls which 
date from the first millennium B.C. and were 
FIGURE 158. Gold plate depicting found in the vicinity of Mingechaur. There is 

a primitive bull from the Maikop a marvellous representation of a bellowing bull 

buriaKactual size) ^^ g^ g^i^ Signet from the Chertomlyk burial 

in the southern Ukraine. 
To the south and southwest of the Caucasus, Mesopotamia, Syria and 
Egypt produced the wonderful bas-reliefs of the ancient Egyptian, Sumerian, 
Babylonian, Assyrian and Hittite cultures, many of them depicting bulls 



379 




and bull-hunts, from the second and early first millennia [B.C. ] when wild 
bulls were common in southwest Asia and northeastern Africa. Usamah ibn 
Munkidh hunted bulls in Syria as late as the 12th century. 

We could find no reference to the existence or disappearance of В о s 
in the histories of the Caucasus. Consequently, we can only infer its history 
on the Ciscaucasian steppes and on the plateaus of the Lesser Caucasus 
from references to it elsewhere: the ancient Russian chronicles, "The Lay 
of the Host of Igor, " epic ballads ("byliny") and "The Instructions" of 
Vladimir Monomakh (see Dolgikh, 1905, Gromova, 1931, Kolesnik, 1936). 

It is very probable that Bos became extinct on the forest-steppe of 
Ciscaucasia and on the Lesser Caucasus some time earlier than it did in 
Europe. 



383 The difficulties of evaluating paleontological, zoogeographical and 
ecological data to determine species origin and fauna formation are 
significantly revealed in this survey of the appearance, development and 
contemporary status of certain Quaternary mammals of the Caucasus. 

These difficulties are all the greater in elucidating the origins of 
eurytopic ubiquitous species (such as European hedgehog, fox and wolf) 
and several other mass species of the Quaternary (such as deer and bison) 
which have nearly stable distribution areas. Nevertheless, it is apparent 
that the local nucleus of the Caucasian mammalian fauna of the Tertiary 
formed in the Pliocene or even in the Upper Miocene, whereas for the 
most part the ancestral forms of Quaternary species do not occur in the 
strata of these periods. 




FIGURE 159. Representation of bulls on the rocks of Beyuk-Dash south of Baku 

Photograph by author, 1945 



1704 



380 



The development of the ranges of many Caucasian mammals was 
demonstrably dependent upon the formation of Quaternary topography. 

The characteristics of the ranges depend to a lesser extent upon the 
species development of morphological-physiological adaptations and the 
interrelationships of biocenoses. 

The history of the development and disappearance of herbivore ranges 
even suggests that the processes proceeded autonomously within each 
species, each independent of the other; the harassment of predators exerted 
a secondary influence on herbivore distribution. 

Three principal types of ranges characteristic for Caucasian mammals 
can be distinguished from species distribution and ecological 
interrelationships: 
384 1- Caucasian forest and alpine type, associated with Tertiary mesophytic 
forest and mountain-meadow landscapes. Examples: the ranges of moles. 
Promethean vole, pine vole and chamois. 

2. Southwest Asian mountain-desert and mountain- steppe type, 
associated with Tertiary xerophytic landscapes which developed on the 
Caucasian landmass in the arid, hot Cezonoic period. Examples: the 
ranges of Persian and red-tailed gerbil, Asia Minor hamster, steppe vole 
and wild goat. 

3. European-Asian steppe type, associated with steppe landscapes which 
developed on the Ciscaucasian plains in the Pleistocene. Examples: the 
ranges of corsac fox, little suslik and saiga. 

The secondary phase of mammalian development proceeded during the 
Quaternary from the bases of these three ranges and resulted from the 
settlement in the Pleistocene and Holocene of European, central Asian and 
southern Asian species on the Caucasus. The range of central Asian desert 
species is the most isolated and alien to the Caucasus (see Chapter V). 

Data drawn from the Recent and based on paleontological studies of the 
distribution areas and on observations of the ecology of the species under 
discussion provide a broad chronological scheme of the origins of 
Caucasian Quaternary mammals. Three different age (stratigraphic) groups 
can be isolated in this fauna: Tertiary (Pliocene), Pleistocene and Holocene. 

The assemblage of local Tertiary forms (or their ancestors) which 
undoubtedly still existed on the Caucasus in the Pliocene was composed of: 
Insectivora, Chiroptera, Carnivora, Rodentia, Proboscidea, Perissodactyla 
and Artiodactyla — these formed a part of the base of the Holocene fauna. 
The assemblage is heterogeneous in both ecology and origin and is subdivided 
into mesophilous and xerophilous classifications. 

The species of the first — mesophilous — subgroup are local; they 
appeared in mesophytic eastern Mediterranean landscapes, i.e., mountain 
and lowland broadleaf forests and alpine meadows. They are the Caucasian 
mole, trogontherium beaver, European beaver, black rat. Promethean 
vole, Caucasian snow vole, long-tailed snow vole, roe deer and Caucasian 
goat. 

The species in this group are eurytopic and ubiquitous. Pleistocene 
deposits contain remains of related forms: white -toothed shrew, fox, wolf, 
bear, deer, tur and bison. From this group the principal survivors are 
the ubiquitous species, the mountain-forest species and those other species 
which could adapt to the mountain-forest. 



381 



The second — xerophilous — subgroup is composed of thermophilous 
species, endemic to arid, hot habitats of the eastern Mediterranean, 
known in zoogeographical literature as Iran-Asia Minor and Mediterranean 
species. Most of the species migrated to the Caucasian Isthmus from the 
south and their principal ranges lie in the southern and southeastern 
Caucasus and extend to the south of the Caucasus. The others survived 
on the Caucasus from the Pliocene, e.g. , the Georgian macaca. 

Typical of this subgroup are the stone marten, Asia Minor suslik, 
common hamster, Persian gerbil, steppe vole and wild goat. Some cf these 
forms can be found now in xerophytic locations on the Greater Caucasus 
385 and its foothills, and sonie only on the southern borders of Transcaucasia. 
It is very probable that this subgroup comprises relatively eurytopic 
animals, which, however, tend to be more xerophilous and southern in their 
adaptations. Examples are the long-eared hedgehog, tiger polecat 
and migratory hamster; the range of the latter passes through the 
eastern half of the Caucasian Isthmus, extending from the mountain steppes 
and semideserts of northern Iran to the steppes of the Russian Plain 
(see Maps 1, 26, 60). The ranges of some species in this subgroup were 
developed on the Caucasus in the Pleistocene, in some cases to a greater 
extent than in the Holocene, e.g. , the common hamster and other southwest 
Asian forms. In other cases, the ranges developed in a northerly direction, 
i.e. , on the plains of Transcaucasia and even in Ciscaucasia in the Upper 
Pleistocene and as recently as the Holocene, e.g., Asia Minor gerbil, 
kulan and goitered gazelle. 

European-Asian hydrophilous and forest species formed the fauna of 
the Isthmus. The emergence of European forest-type animals on the 
Caucasus from Western Europe was feasible in the Tertiary, the Pleistocene 
and all epochs when the broadleaf forests of Europe and the Caucasus 
were connected through Asia Minor and the Balkans or through the Russian 
Plain. The formation of parallel species, especially in mountain forms 
like lynx, Caucasian snow vole, roe deer and chamois, was also possible. 

Those species which penetrated the Isthmus from the north, southwest 
and southeast belong to the younger — Pleistocene — genetic group. The 
reasons for their settlement lie in the ancient formation of the landmass 
of the northern Caucasian Isthnius, in the development of forest and 
meadow-steppe landscapes during the time of climate -cooling, in the 
development of steppe and semidesert landscapes in the Ciscaucasian plains 
during the time of climate -warming. It is very probable that there were 
two migratory routes to and from the Caucasus open to forest species in 
the Pleistocene — a northern one along the Don floodplain, and a southwestern 
one along the northern coast of Asia Minor. 

By either of these ways, the elk and the typical European brown bear 
could have penetrated the Caucasus. Only the northern route would have 
been feasible for the striped field mouse, the large Asian roe deer and 
other species; pine marten and European wildcat are more likely and 
common red-backed vole is certain to have followed the southwestern route. 
The time of their emergence on the Caucasus probably varied for each of 
these species — common hamster penetrating a little earlier, European 
brown bear and Asian roe deer a little later. 

The Pleistocene assemblage also included steppe elements: corsac fox, 
little suslik, saiga and other species which emerged on the Caucasus in 



382 



the Lower Pleistocene, and steppe lemming which emerged later. In the 
Pleistocene, the northern steppe influence reached even into eastern 
Transcaucasia. 

The development of southern southwest Asian species was difficult in 
the Pleistocene because of the progressive cooling. 

The so-called interglacial epochs, warm, dry periods, can be traced 
on the Isthmus from the distribution areas of Turan desert species and, 
to some extent, of southwest Asian species. 

The latest genetic group to emerge on the Isthmus in the Holocene was 
composed of southern, thermophilous species which migrated to the 
Caucasus during the postglacial warming and dessication of landscapes. 
386 It is comprised of jackal, striped hyena, lion, tiger, jungle cat, kulan, 
goitered gazelle and probably some Insectivora and Chiroptera. Some of 
these species became extinct on the Caucasus as the result of human 
activity. 

The Norway rat and the European races of house mouse can be counted 
as casual species, which were brought by ocean and river vessels of the 
ancient Greeks, Slavs, Khazars and Varangians, and later by railroad 
and automobile. 

A study of the characteristics of former and contemporary ranges of 
Caucasian mammals shows the effects of a great natural dynamic on the 
populations and on the range patterns of isolated species during the 
Quaternary. It also reveals the significant relationship between the 
anthropogenic influences of the historical epoch and the populations and 
ranges of many species. 

The discontinuities in the ranges of stenotopic and relatively stenotopic 
manamals of the Caucasus can be explained in the majority of cases by the 
alternation of dry and humid epochs during the Cenozoic and the orogeny 
and glaciation of the mountain ranges. 

The most characteristic are the high-mountain breaches in the ranges 
of Caucasian endemics — Caucasian snow vole. Promethean vole and 
chamois — in the region of the Surami Range; their age is probably Lower 
Holocene. Discontinuities in the ranges of mesophilous and forest species, 
e.g., mole, pine vole and roe deer, are peculiar to central Ciscaucasia 
and eastern Transcaucasia; some date from the Pleistocene and some from 
the post-Glacial. Among the xerophilous species, breaks can be observed 
in the ranges of common hamster, steppe vole and wild goat. 

An analysis of discontinuities like these, particularly those occurring 
in the ranges of rodents, reveals the patterns of two xerothermic epochs 
on the Caucasus — one in very ancient and the other in Recent time. The 
effects of a cold, humid epoch between these two can also be traced. 

Chapters IV and V are devoted to a general analysis of the distribution 
of all Holocene animals, based on zoogeographical maps, and to an attempt 
to evaluate with some precision the degree and kind of faunistic influences 
of the adjacent territories through a study of the stratigraphic and 
geographic variabilities. 



383 



387 Chapter IV 

STRATIGRAPHIC AND GEOGRAPHIC VARIATION* 
IN CAUCASIAN QUATERNARY MAMMALS 

The study of the phylogeny of Caucasian Quaternary mammals presents 
many problems. 

The fossil remains from most localities are fragmentary, and suffer 
from post mortem crushing, making biometric studies very difficult. Skull 
material and skins of Recent small animals have been collected in different 
years and seasons by many workers using a variety of preservation. For 
these reasons conclusions on the spatial distribution and evolution of animal 
species are not always reliable and convincing. For example, the annual 
and seasonal changes in growth rate and weight among rodent populations 
with short life cycles (e.g., moles and voles) may be more pronounced 
than variation which is related to local environmental changes. 

Insectivores and murids preserved in spirit and by dry method are of 
little use for detailed systematic study, as pointed out by Shidlovskii 
(1953a, b, 1954a, b). Data on geographic variation in fur-bearing animals 
is mostly based on information from the fur-trading stations and must be 
accepted with reservations, as a study of this type must take into account 
individual, sex and age variability within the populations. 

It is felt that comparative study of the morphology and systematics of 
the mammals of the Caucasus and the Russian Plain may contribute to an 
understanding of the development of the fauna, and therefore a summary 
of data on some orders is presented below. Most of the work on the evolution of 
the species through time has been based on better material, more amenable 
to biometric study. 



Order INSECTIVORA 

The number of fossil insectivores in the Caucasus is negligible. The 
Middle Pleistocene hedgehogs (Hemiechinus aff. auritus and 
Erinaceus aff. europaeus) from the bitumens at Binagady on the 
Apsheron Peninsula are practically indistinguishable from the Recent forms 
in size and structure of the mandibles and teeth. 

The geographic variation of the common hedgehog in the Caucasus is 
reflected in the darker spines and abdominal fur in gray woodland and 
wooded lowland populations . Dark varieties — Erinaceus europaeus 
concolor, E. europaeus ponticus — occur on the Black Sea 

• Because it seems to us more correct to refer to evolutionary and geographic changes in living organisms as 
variation we reserve the more commonly used term variability to denote individual and age changes . 



384 



388 



(385' 



coast of the Caucasus and in Asia Minor, the Eastern European 

E. europaeus rumanicus is found in Ciscaucasia, and the light - 

colored E. europaeus transcaucasicus in the semideserts of 

Transcaucasia. 

The variation in the Recent long-eared hedgehog is slight over the 
Caucasian Isthmus. The populations of the small forms, Hemiechinus 
auritus calligoni and H. auritus brachyotis, occur in the 
semideserts of eastern Ciscaucasia and in the middle Araks valley. 

Detailed information on the stratigraphic variation of the Caucasian 
moles (genus Talpa) is not available. The Recent mole populations 
can be subdivided into distinct geographic varieties, particularly in the 
areas which became isolated long ago. 

The biggest Caucasian moles (Talpa caucasica ognevi) inhabit 
the southern slopes of the western half of the Greater Caucasus and the 
Trialet ridge (Ognev, 1926a, 1928; Stroganov, 1948)*. The moles decrease 
in size to the south and southeast. Populations of the smallest varieties, 
T. orientalis transcaucasica and T. orientalis 
talyschensis, occur in the ridge s of the Armenian Highlands and in the 
forests of Talysh. These populations are isolated from the main area of 
distribution of the species; however, N.K. Deparma has recently found 
a relatively small population of moles in the southeastern Caucasus, even 
smaller than the Talysh population. 

The size of the skulls and metatarsals of the Caucasian moles is given 
in Table 75. 

TABLE 75. Geographic variation in size (mm) of Caucasian moles* 





ex > 


П) 




a. 






-o ■;; >" 





'^ 


« с 


'^ 


Measurements 


ter an 

sus T 

ogn 


asus 
cast 

[IS 




3 - 


4-" 

С ." 
















f Gre 
Cauc 
asic 
imen 


5 3 M 

tt « (0 <u 


Cauc 
a or 
tali 
ecim 


Cauc 

ta li 
asic 
cime 


a or 
che 
imen 




" >-. о " 

4.^ (U " u 
m m Э Q, 


2 a, й- 


1-, d. >- Оч 

1) Л^ « " 


b.. '- lU 

^ (u g a. 


i^i^i 




4; S « "■ 


M « П) ^ 


S "■ t- l^- 


S >-i <e CO 


ca "J rt " 




Э J 00 


H CN 


J H .-( 


»j ^D 


H H .- m 


Condylobasal length 


35.9 


33.9 


30.1 


32.4 


29.6 


of skull 


35.1-37.2 
14.5 


33.1-35.7 

13.4 


29.8-32.6 
12.2 


31.3-33.0 
11.2 


29.5-29.8 


Length of upper tooth 


11.1 


row 


14.0-15.0 
19.6 


12.5-13.8 
18.3 


11.9-12.7 
16.9 


10.5-11.7 
16.5 


11.0-11.2 




15.5 


Length of foot 












19.0-20.0 


17.5-19.2 


16.0-17.9 


14.2-17.5 


15.0-16.0 



Note. Mean value of observed ranges in the numerator, limits in the denominator. 

* Measurements by Stroganov (1948), Dal' (1944a), Vereshchagin (1945b) (Talysh). 

This size decrease to the southeast may be attributed to the prolonged 
isolation of moles on mountain ranges under conditions of increasing 

* In 1948 S.U . Stroganov tentatively identified 8 specimens of moles from the region of Kutaisi and 
Borzhomi as the Italian mole T.romana ognevi. Subject to the final clarification of affinities 

of the Caucasian and west Mediterranean moles, we regard this form as a subspecies of T. caucasica; 
T.europaea transcaucasica, described by Dal' (1944a) from the Pambak ridge, is tentatively 
referred to T. orientalis. 



385 



draught, increasing annual temperatures and depletion of food resources, 
i.e., land invertebrates. 

The speciation of the Caucasian shrews can be understood in the light 
of the fact that these mesophilous forms, like the moles, were isolated for 
a long period in the forest -mountain massifs. 
^^^^ Recent shrews of the genus Sorex on the Caucasian Isthmus (e.g., 
Sorex minutus) also decrease in size from west to southeast. The 
Caucasian population of the common shrew is at present regarded as a 
subspecies of the European shrew, S. araneus satunini (Ognev, 1928). 
This shrew is distinguished by a short, wide nasal region and shorter 
mandible, etc. 

There are, however, certain larger local species of shrews which have 
evolved in the Caucasus, for example, the darker-colored and long-tailed 
S. raddei, closely related to the European S. araneus. 

European water shrews in the Caucasus (genus Neomys) are 
morphologically and ecologically more specialized than land shrews, but 
their genetic relationships are not well known. The forms described — 
Neomys le pt о d a с t у lu s , N. s с h e Ik о vn i ко v i, N. balcaricus 
and N. dagestanicus — are regarded here as subspecies of the European 
N. fodiens. The first two of the above are characterized by narrow 
paws, as pointed out by Satunin (1915a), whereas the Dagestan 
subspecies is characterized by a poorly-developed tail, pale coloration, etc., 
characters typical of animals inhabiting lands with progressive desiccation 
(Spain, the Crimea). 

The structure of the skull and teeth of the white -toothed shrews of the 
Middle Pleistocene from the Binagady burial (Crocidula russula, 
C. leucodon) does not differ significantly from the contemporary forms 
inhabiting eastern Transcaucasia. Southern white-toothed shrews are much 
more widely distributed over the Caucasian Isthmus than other species; 
particularly abundant is C. russula gii Id e n s t ae dt i which inhabits 
both the dry and the humid zones. 

Geographic color variation is well seen in the Recent species and is 
directly related to the aquatic biotopes in the hot semidesert and humid 
shady forests of Colchis and Asterabad. According to Shidlovskii (1953b), 
the long-tailed shrews of Transcaucasia are smaller and darker in the 
humid forests of the coastal plains of the Black and Caspian seas than in the 
390 center of the country and in the dry areas of easternTranscaucasia (Table 76). 

Clearly, if the isolation of the Caspian shrew can be proved, then any 
similarity in color will be the result of convergence due to the ecological 
similarity in the areas of Colchis and Asterabad. 



Order CHIROPTERA 

With a few exceptions, fossil Chiroptera are known in the Caucasus only 
from the Holocene beds. Thus there is no basis for discussion of the 
phylogenetic changes in tiiis order. According to the data available in the 
literature (Bobrinskii, Kuznetsov, Kuzyakin, 1944; Kuzyakin, 1950), the 
geographic variation in some Caucasian Rhinolophidae is expressed in the 
development of lighter coloration from northwest to southeast and in the 
increase in body size with increasing altitude above sea level 
(R. h i p p о s i d e r о s , R. ferrum equinum). 



386 



The widely distributed genera Myotis and Vespertilio are 
sometimes intermediate between the European-Siberian and central Asian 
populations (e. g. , Myotis mystacinus), while other species (e.g., 
Vespertilio serotinus) are closer to the European forms. 

TABLE 76. Geographic variation in dimensions (in mm) and coloration of long-tailed shrew in Transcaucasia* 





Western Transcaucasia, 


Central Transcaucasia , 


Eastern Transcaucasia, 




Colchis 


Kura valley 


Talysh 


Dimensions and color 


Crocidura russula 


Crocidura russula 


Crocidura russula 




monacha 


guldenstaedti 


caspica 




150 specimens 


170 specimens 


40 specimens 




70.6 


74.1 


70.6 


Body length 


60-89 


60-91 


63-80 




12.7 


13.0 


13,1 


Foot length 


11.0-15.0 


11.0-15.0 


12.0-13.6 


Condylobasal length of 


18.1 


18.7 


19.3 


skull 


16.7-19.4 


17.5-19.8 


18.6-20.2 


Length of upper tooth 


8.1 


8.2 


8.9 




7.6-8.8 

Whole body dark 
chestnut brown, tail 


7.3-8.9 

Upper part of body gray- 
brown, lower part 


8.3-9.4 


Color 


Whole body dark chestnut 
brown, tail uniformly 






very dark all over 


grayish white; tail 
uniformly dark 


dark 



Note. Mean value of observed ranges in the numerator, limits in the denominator. 

• According to Shidlovskii (1953b), whose measurements are given in the above table, C. russula 
caspica must be identified as a distinct species, С caspica. This, however, needs further study, 
since the long-tailed shrew is a highly variable species, both ecologically and morphologically. 

391 In order to reach reliable conclusions on the geographic variation in the 

Caucasian bats, good southwest Asian data must be studied, as the ranges 
of distribution of most of the species extend from the Caucasus far to the 
south and southwest. 



Order CARNIVORA 

The paucity of carnivore remains in Quaternary burials causes difficulties 
in the study of the morphological evolution of this order. The only exception 
is the Binagady locality, where there are particularly numerous canid 
remains. 

The Middle Pleistocene Caucasian wolves are characterized by a smaller 
brain cavity, narrower nasal cavities, a narrower cheek region, shorter 
upper tooth row and smaller intertemporal width (Figure 160). It is of 
interest to note that no differences were recorded in the shape and 
structure of the teeth themselves. The decrease in the length of the upper 
tooth row was due both to closer spacing of the teeth and to the gradual 
disappearance of M^. In 33 wolf skulls from Binagady, M was present 



387 



н 

t3 

12 
11 
10 
9 
8 
7 
6 ■ 

5- 

А - 
3 ■ 
2 




in only 2 (i.e., 6%). Out of 222 
skulls of Recent wolves from the 
U. S. S. R. , M^ was found in only 
0.9%. Changes in the limb bones 
are reflected in the gradual 
narrowing of the scapulae with time 
(Vereshchagin, 1951b). 

The present geographic variation 
of the wolves on the Isthmus has 
been poorly studied; however, it is 
known that the wolves of the 
Transcaucasian plains are somewhat 
smaller than those of Ciscaucasia, 
while the Armenian Highland forms 
are larger. According to Dal' 
(l951a), the Armenian Canis 
lupus hajastanicus is no 
smaller in size than the Kuban wolf, 
and is distinguishable from it by its 
long, soft fur; its back is bright 
whitish gray and yellowish with a 
black tint. The fur traders in the 
Caucasus distinguish the "steppe", 
relatively light-colored wolves of 
the open plains, and the darker 
varieties from the "forest". Wolves 
of bright iron-rust color occur in 
Kabarda. This is a fairly rare 
example of an aberrant variety. 
Three skins of this variety are in 
the collections of the ZIN Museum. 
The probable Middle Pleistocene ancestor of the Recent fox, Vulpes 
khomenkoi from the fossiliferous beds in eastern Transcaucasia, is 
characterized by small molars and short canines, like those of the Arctic 
fox (Bogachev, 1938c; Vereshchagin, 1951b). Towards the end of Middle 
Pleistocene times (Khazar stage) the fox evolved into the modern form and 
the subsequent changes in the structure of the skull have been negligible. 
^^^ The cranial features of the Middle Pleistocene foxes of eastern 

Transcaucasia (Binagady), however, are quite variable. This variation is 
characteristic both of the Recent fox of the Kura Lowlands (Vulpes 
vulpes alpherakyi) and the foxes of the highlands — V. vulpes 
kurdistanica, V. vulpes alticola. The basic structure and size 
of their teeth, however, does not differ fronn the former species, as shown 
in the graphs (Figure 161). 

The skull size and fur of the Recent Caucasian foxes display a marked 
geographic variation. The general pattern of variation has been traced on 
individual skulls and pelts by Ognev (1931), who published a distribution 
map of the described subspecies of Caucasian foxes: Vulpes vulpe.s 
stepensis, V. vulpes karagan, V. vulpes caucasica, 
V. vulpes alticola, V. vulpes kurdistanica, V. vulpes 
alpherakyi. 



12 16 20 2U 28 32 36 40 44 "" 

FIGURE 160. Variation in the temporal index (ratio 
of the intertemporal width to the length of the brain 
case expressed as a %) of Middle Pleistocene wolves 
from the Caucasus 

1 — Recent Canis lupus cubanensis; 2 — 
Fossil C. lupus binagadenis, eastern 
Transcaucasia, Binagady. Dots indicate mean 
values 



388 



The distribution of the Caucasian foxes, based on the differences in the 
fur coloration (material studied at fur-trading posts), has been accurately- 
described by Kuznetsov (Bobrinskii, Kuznetsov, and Kuzyakin, 1944). 

The first distribution map of local variation (in such features as fur 
color and other associated characters) and size variation in the Caucasian 
foxes was published by us (Vereshchagin, I947d). 

Pelts obtained in the course of hundreds of years are subdivided into 
six races by the local fur traders: Ukrainian, Don, Kuban, North Caucasian, 
Yerevan and Transcaucasian. 

Our study of the pelts of more than 13,000 specimens of foxes at the 
Rostov and Tbilisi fur stations may be briefly summarized as follows. 
393 Ukrainian. This race is subdivided into north and central Ukrainian. 
These are big foxes with relatively stiff guard hair, rusty white on the 
back and straw on the sides. The belly is mostly rust in color and 
the upper surface of the paws is black. This variety occurs, though rarely, 
in the Ciscaucasian plains. Populations of these foxes occur in pockets 
among the typical north Caucasian foxes, from the northwestern Manych 
region to the Kabarda plains. According to the fur traders, the presence 
of this variety in the 1940's in Ciscaucasia was noticed only after World War II. 

Don. Fairly large light-colored foxes, rusty white, with red shoulders 
and belly. The guard hair is shorter than in the Ukrainian race. The most 
commonly occurring colors are probably related to age and sex: rust, light 
rust, rust gray; more rare are dark rust and gray rust specimens. 

These foxes lived round the coast of the Sea of Azov, the lower reaches 
of the Don, and the Sal'sk steppes. In the east their range extends to the 
Yegorlyk, in the south, to the Yeya. 



[392 



f 
I 
I 



Д 

/\ 
/\ 
/ \ 
/ \ 

/ \ 
-» \ 
\ 



\ 2 A 
\ у \ 



:Awv\ 



56 58 60 62 64 66 68 13 /4 15 16 




10 11 12 13 



FIGURE 161. Variation in tooth size of Transcaucasian Middle Pleistocene foxes 

I — length of lower molar row; II — length of crown of М^; III — length of crowns of M2+M3; 
1 — Recent Vulpes vulpes a Iphera ky i; 2 — fossil V. vulpes aff. a Iphera ky i; 
eastern Transcaucasia, Binagady. Dots indicate the mean values 



389 



КиЬал. These foxes are somewhat larger in size than the Don race. 
The characteristic bright rust and even reddish color on the face, back 
and, particularly, neck and shoulders produces a noticeable "cross" 
pattern. The guard hair on the back is shorter than that of the Ukrainian 
race. Individuals of very rich rust color are most common, while rust 
gray individuals are less abundant, and even more rare are the light rust 
forms. 

The range of this race is on the steppes of the Kuban region, on the 
Trans -Kuban inclined plain from the Kerch Strait in the west to the Kuban- 
Kuma water divide in the east. 

North Caucasian. These are foxes of a somewhat smaller size than those 
of the Kuban race. The fur is coarse; on the back the guard hair often 
forms "curls" — patches of standing coarse hair. The belly is black to 
dark gray. These foxes occur in two varieties, the red and the paler red- 
gray forms, the latter becoming more abundant to the southeast. The 
foxes identified as V. vulpes с au с a s i с a belong to this variety. 

The distribution of this race is from central and eastern Ciscaucasia, 
from the Trans-Kuban Plainto the Caspian coast, and from the lower Kuma 
and Terek to the Samur. Pelts of such foxes also come from eastern 
Transcaucaia and show great similarity in coloration and other features 
with the eastern Ciscaucasian form. 

Transcaucasian. This race is represented by the small foxes of the 
eastern Transcaucasian plains. They are subdivided into three varieties: 
Transcaucasian red, Transcaucasian red-gray, and Transcaucasian gray. 
Their fur is relatively coarse and short, often with considerable admixture 
of black hair on the rump and shoulders. The foxes are characterized by 
a yellowish rusty "band" of variable width, which runs along the length of 
the back. The band is particularly conspicuous in the paler colored young 
specimens; as a rule, young individuals are gray in color. The small fox 
of eastern Transcaucasia, described as V. vulpes alpherakyi, 
belongs to this race. 

Foxes of the Transcaucasian race inhabit the lower reaches of the Terek 
and Sulak, the western coast of the Caspian, the plains of eastern 
394 Transcaucasia and the middle Araks valley. In the forties, the Transcaucasian 
Vulpes vulpes к a r a g a n, disseminated in the fox population of eastern 
Transcaucasia, was identified as belonging to this race. 

By the state standards of the forties, this is a small fox with soft 
uniformly dark gray and even sonnewhat dark brown fur. The limbs below 
the elbow and knee joints are black or black-brown. This variety has little 
in common with the central Asian karagan fox, as the former is an 
aberrant form, a dark-colored variety of V. alpherakyi, like the black- 
brown varieties of the arctic fox. 

Yerevan. A large fox, though smaller than the Ciscaucasian fox, with 
a weakly ossified skull. It is clearly distinguishable from other Caucasian 
varieties by its extremely thick and soft fur. The guard hair is very long 
and silky to the touch, and the color varies from light yellow to a dirty 
rust brown. Foxes identified as V. vulpes alt i со la and V, vulpes 
kurdistanica belong to this race. 



390 



(395; 



psipnjs 




<N 


о 


о 


35 


IN 


ее 


СП 


CD 


■* 


ю 


ОС 




1 


о 


Ю 




lO 






со 




г- 


о 


00 


СО 


(N 


suT>is }o jaqmnM 


со 


СО 


со 


<м_ 


- 


00 


ю 






IT- 
CO 


со 


с- 

со" 




laiSuoiAi 


1 


1 


1 


1 


1 


1 


1 




U0 


1 


1 


1 


1 


XbjS Ixoj ssoj3 


1 1 1 1 1 1 1 1 1 <N 1 1 1 


ивлэ1эл 


1 


1 


1 


1 


1 


1 


о 
IN 


(М 


со 
со 


^' 

со 


1 


eg 


ОО 
(N 


иБ§ИВ>) 


1 


1 


1 


1 


1 


1 


1 


1 


1 


ю 

(N 


1М 


1 


1 


XbjS UBiSBotiTOSUBii 


1 


1 


1 


(М_ 


1 




г- 


1 


1 


CD 




о 
о 


СО 














со 










со 




1М 


XbjS 


1 


1 


1 


1 


1 


U0 


1 


1 


1 


^ 


U0 

оо" 


1 


1 


-рЭ1 UBlSTOnBDSUBJX 




















1М 




















о 








СО 




со 




РЭ1 UBlSBOnBOSUWl 


1 


1 


1 


1 


1 


оз 


1 


1 


1 


ю 


1 


ю 


со 


ХшЗ 




























-paj ивтзвэпЕэ quoM 


1 


1 


1 


1 


о 


со' 


со 

из 


1 


1 


'^ 


^ 
N 


со 


о 






о 




ю 


о 


ю 












(N 




psj UBTSBoriBO quoN 


1 




ю 


оо 

00 




U0 


о 


1 


1 


(N 


1 


О 


со 

(N 






■* 




со 


ю 


















UBqnM 


Ю 






о 


(М 


























ю 




















UOQ 


с- 






го 
























(М 




со 


со 


















UBlUTBJ>ln 


о 


<Х> 




(М 


о 


1 


1 


1 


1 


1 


1 


1 


1 








са 




























Р>ч 




























л 




























^ 




• 






• 


._, 


















с 












м 












с 


00 




J3 




л 


^ 






с 


и 




С 






о 

s 


> 




J 




■о 
о 


П1 

с 

3 




>> 


(Л 


2 
н 


X) 


> 


с 


о 
и 


с 
> 


> 

о 
2 


'во 

N 




СП 

с 


с 


о 




се 


J3 

3 


О) 




e 






с 




а. 






с 








с 


л 


"о 


> 


00 


а. 
с 


а. 




DO 

о 


U 


с 


3 

ш 
о 


3 
О 


ел 
о. 


00 

X 
с 
2 

с 


се 

2 




а. 


о, 
<и 

с 

nj 

э 


3 

1^ 


J3 


Е 

3 


о 


о 

и 

с 


а! 

.а 




2 

н 


со 

с 


се 

> 




(Л 


с 
2 

н 


С 
2 




о 
о 


а. 

се 

и 


о 

"о 

и 


1-4 

С 

и 

и 


с: 
<и 

и 


п! 

ао 

3 

2 


1) 

6 

< 


се 





391 



i,dyb JxABLE 78. Geographic variation in size (cm) of fresh fox pelts from the Caucasian Isthmus and the Ukraine 



Races and fur station 



Length froni 

nose to tail 

base 



Middle 
width 



Area of 
pelt (cm^) 



Length of tail 
without fur 



Number of 
skins measured 



Central Ukrainian, Stalino 
Don, Novocherkassk . . . . , 



Kuban, Slavyanskaya .... 
Nonh Caucasian, Blagodarnoe 
Mongrel, Batumi 
Mongrel, Tsalka 
Transcaucasian, Yevlakh 

Transcaucasian, Baku .... 
Yerevan, Tsalka 



Yerevan, Yerevan 



85 



79-90 

76 
73-86 

79 
71-94 

76 
70-86 

77 
70-88 

76 

70-84 

68 
66-80 

66 
54-76 

75 
70-84 

77 
70-90 



24 



23-25 

22 
21-26 

22 
19-25 

21 
19-24 

14 
14-17 

15 
15-17 

16 
16-22 

15 
13-22 

15 
13-18 

20 
17-21 



4106 



3384 



3380 



2174 



2312 



2208 



2020 



2302 



3048 



42-47 

42 
39-45 

44 
37-45 

41 
35-43 

39 
37-43 

40 
40-44 

38 
38-44 

34 
24-45 

42 
35-45 

43 
40-48 



31 



34 



30 



28 



28 



34 



36 



65 



23 



29 



Note. Mean value in the numerator, observed range in the denominator. 



397 



These species inhabit the Armenian highlands, and probably the ranges 
of northwestern Iran and eastern Turkey in the south. Similar though not 
identical skins come from the highlands of the Greater Caucasus, 
particularly from south Ossetia and central Dagestan. 

It should be mentioned that the age variation of the animals could not be 
taken into account in the construction of the distribution map of the fox 
races (Figure 162) and the diagram of the distribution of different color 
types (Figure 163). The distribution and relative proportions of the different 
races are given in Table 77. 

Unusual coloring is observed occasionally in certain strains of Caucasian 
foxes, e. g. , the "sivodushka", which has dark blue -gray to black fur on the 
belly and blackish brown fur on the back. Another type is the cross fox, 
which has reddish yellow background fur with a dark brown cross extending 
from the nape of the neck to the tail. This form occurs occasionally in 
Ciscaucasia and Transcaucasia. Peculiar silver foxes with chocolate-purple 
fur are occasionally found in Dagestan. Silver foxes rarely occur in the 
Zakataly and Nukha regions in Transcaucasia, only 1-2 specimens being 
found in 20,000 animals. 



392 




393 




394 



(396). 



TABLE 79. Geographic variation of skull dimensions (in mm) of foxes from the Caucasian Isthmus' 



Regions and collection sites 



asic length of skull 



Width between ends 
of auditory canals 



Number of skulls 
studied 



Western Ciscaucasia and Don area 
Tarasovka area 



Azov area 
(^^'^/Sal'sk area 



Trans-Kuban Plain, Maikop 



Central Ciscaucasia 



Kursavka area 



Eastern Ciscaucasia 

Ipatovo, Beshpagir 

Ordzhonikidze 

Kizlyar, Makhachkala 



Greater Caucasus 



Gvilety 



Black Sea coast 



Tuapse-Gagry 



Eastern Transcaucasia 
Kirovabad 



Lesser Caucasus 



Borzhomi 



Yelenovka 



131.5 
125-138 

128.0 
125-137 

129.0 
123-138 

133.5 
128-142 



133.0 
124-141 



134.5 
123-150 

132.0 

123-141 

129.0 
120-143 



133.0 
122-138 



128.0 
126-135 



121.5 
115-129 



125.0 
118-133 

128.0 
119-138 



47.0 


46-49 


46.8 


46-48 


47.0 


45-49 


47.5 



46-51 



47.5 
46-50 



47.0 
46-49 

48.5 
46-51 

46.9 
46-49 



47.7 
46-50 



45.8 
43-48 



44.4 
43-46 



45.0 
44-46 

47.2 
44-50 



10 



37 



48 



11 



14 



Note. Mean values in the numerator, observed ranges in the denominator. 

* Material used is in the ZIN and Moscow University collections. Skulls used for measurements were those 
of mature males and females, with replaced teeth or teeth in early stages of wearing. 



395 



Fur traders do not subdivide the fox skins from the western plains of 
Transcaucasia. into races. These foxes are small and their fur is the 
398 usual type of rust in color. As far as the pelts at the Batumi station show, 
rust gray foxes predominate in Adzharia; they amount to 71 % of all the 
foxes. Foxes similar in color and size to the Transcaucasian race number 
16%. Foxes of the type transitional from the Yerevan to Transcaucasian 
occur on the margins of the Dzhavakhetia Highland (Bogdanovka, Tsalka). 

Dimensions of skins and skulls of foxes of some of the races mentioned 
are given in Table 78 and 79. 




FIGURE 164 

Analyses of all the data on the distribution of the coloration types and 
dimensions of skins and skulls produce a very complex picture of the 
geographic variation of foxes on the Caucasian Isthmus. It is clear that 
on the Ciscaucasian plains the foxes become consistently smaller and 
paler in color from west to east. In Transcaucasia the altitude factor is 
superimposed on this regularity. Nevertheless, the smallest and least 
brightly-colored foxes are characteristic of the easternmost populations 
inhabiting the hot Kura-Araks Lowland. Another example of the connection 
between fur color and landscape are the foxes in eastern Transcaucasia. 



396 



Data on their distribution were obtained from 41 fur stations for the 1935/36 
season. The total number of pelts used in the construction of the diagram 
(Figure 164) was 28,250. - 
399 From the review of the diagram it is possible to conclude that the 

palest-colored foxes occur in the zone of semideserts. Westward, towards 
the foothills and mountains, the abundance of bright -colored specimens 
increases. 

In the Zakataly forest -mountainous region, 4.2% of the furs are silver 
foxes, 4.2% sivodushka, and 2.1% cross foxes. 

It may be mentioned that to the south and southeast of the Caucasus are 
located the areas of distribution of the small races of the same polytypic 
group of the Palearctic black-eared foxes: V. vulpes f 1 a ve s с e n s Gray, 
V. vulpes persicus Blanf . , V. vulpes leucopus Blyth. 

Taking into account such taxonomic variability of the Caucasian foxes 
and following the generally accepted criteria for establishing the area of 
origin of the species, the subprovince of the east Mediterranean and 
southwest Asia is the area from which the Holarctic fox most probably 
originated. 



••1 

' \ 

I \ 

! \ 



\2 
\ 
\ 
\ 
\ 



18 




W 
14 


12 


A ' 


10 


1 . 


6 


1 


6 


W 


и 


■ 'A 


г 


• 4V 




1 I.I 



f. 


. in \ 

h 

1 \ 
/ \ 


-:^XJ. 


1 1 1 1 1 



42 44 46 48 50 52 



11 12 13 



5 6 7 8 9 



FIGURE 165. Variation in size of teeth of corsac fox in the Middle Pleistocene 

I — length of lower molar row; II — length of Mi crown; III — length of M2+M3. 1 — Recent 
Vulpes corsac, Central Asia; 2 — fossil V. aff. corsac, eastern Transcaucasia, Binagady. 
Dots indicate the mean values 



The evolution of the skull of the corsac fox was much more pronounced 
than in the common fox. The remains from the Binagady asphalt indicate 
that the Middle Pleistocene east Caucasian corsac fox was somewhat smaller 
than the Recent Ciscaucacian and central Asian varieties (Vereshchagin, 
1951b). In addition, the size of M2 and Мз has considerably decreased, 
and Мз has even completely disappeared since the Middle Pleistocene 
(Figures 165, 166). The normal sized Мз was present in the Binagady 



For the sake of brevity the numbers were not tabulated. 



397 



corsac fox (43 cases), whereas in the Recent species this tooth is 
present in 66% (8 out of 12 cases). The Recent corsac foxes increase 
in size to the east, in their presumed country of origin, i.e., the 
steppes of the Baikal region and Transbaikalia, where they attain their 
maximum size . 

Brown bears in the Caucasus have noticeably decreased in size since 
the Middle Pleistocene. The Binagady bear from the Middle Pleistocene 
beds of the Apsheron Peninsula is closely related to the Recent large 
Caucasian bear, Ursus arctos caucasicus, but the Binagady species, 
which is probably ancestral to the Caucasian bear, is characterized by 
400 bigger molars of a more carnivorous type. This is particularly noticeable 
in the shape of the sharp pointed heel of mI. The decrease in the size of 
the animals and their teeth towards the Holocene was accompanied by the 
flattening of the grinding surface of the teeth with the development of 
herbivorous habits. The Binagady Pleistocene bear may actually be 
identified as a distinct species (Vereshchagin, 1951c). 

The biggest was the Middle Pleistocene bear of the Russian Plain, 
known from the Lower Kama (Mysy). This bear, a contemporary of the 
Binagady bear, has been identified by us as U. kamiensis N. Ver. sp. 
nov. The skull is elongate with a low forehead (Figure 167). In size (basic 

skull length 381, 386 mm) it was no 
smaller than the cave bears of the 
Pleistocene. In the Upper Pleistocene 
the bears of the Kama area (Tatar 
A. S. S. R. ) began to decrease in size 
and to develop a convex forehead. The 
latter feature is particularly noticeable 
in the skulls of the bear 
U. к a r m a 1 к i e n s i s N. Ver. sp. 
nov. (Figure 167, 2) from the asphalt 
near the village of Nizhnie Kalamalki on 
the Shemsha rivulet. In postglacial 
time the bears of the Russian Plain 
and the Caucasus decreased 
considerably in size, as confirmed 
by the size of the last molars. Our 
diagram (Figure 168), which combines 
the elements of the stratigraphic, 
geographic and individual variability 
in the size of M^ of bears, illustrates 
these facts. 

The Recent populations of bears 
of the Caucasus are genetically highly 
heterogeneous. In the present case 
it is probably impossible to speak 
of the regularities in the geographic distribution of one monotypic species 
since, according to Smirnov's (l9l6a) studies, the region is inhabited by 
the big Caucasian subspecies and two small races — the northern and the 
southern, or Mediterranean. The interbreeding of these three forms (the 
northern race has probably migrated to the region relatively recently) 
makes the studies of the geographic variation of the species very difficult. 




FIGURE 166. Lower jaws of corsac foxes 

1 - Recent, Central Asia, No. 9470, ZIN; 

2 — fossil, eastern Transcaucaasia, Binagady, 
No. 23674, ZIN 



398 



There is a high degree of variability in skulls of badgers from the Middle 
Pleistocene of Binagady. Skulls with features of both the sand badger (Me le s 
meles leptorhinus) and the common badger occur at the same locality. 
Most of the badgers, however, were closest to the Recent badger of eastern 
Transcaucasia and northern Iran — M. meles minor and M. meles 
canescens Blanf. It is interesting that the brain of the Recent badgers 
has become smaller, which attests to a decrease in metabolic rate 
(Figure 169). The first upper molar of the badgers from eastern 
Transcaucasia has decreased 8% in width, relative to length, since the 
Middle Pleistocene. However, badgers from the Bronze Age of the Armenian 
Highland had a wider M than the early Apsheron forms (Vereshchagin, 
1951b). 



401 





FIGURE 167. Skulls of bears 

1 — Recent Ursus arctos caucasicus, Greater Caucasus, No. 6169, ZIN; 

2 - U. karmalkiensls N. Ver. sp. nov, Tatar A.S.S.R., Nizhnie Karamalki, Upper 
Pleistocene, No.3, Kazan University; 3 — U. kamiensis N. Ver. sp. nov., Tatar 
A.S.S.R., Mysy, Middle Pleistocene, No.l, Kazan University 



399 



The Recent badgers of Ciscaucasia are larger in size, while those from 
Transcaucasia are the smallest. 

Skulls of Vormela pere gu s n a of eastern Transcaucasia have not 
altered appreciably since the Middle Pleistocene. The Recent tiger 
polecat decreases somewhat in size from the west to east in the steppe 
and desert zones of the U. S. S. R. 

The Upper Paleolithic gluttons from western Transcaucasia (Figure 57, 2) 
differ from the Recent tundra-taiga species in the deeper mandibles and 
larger molars, in this respect exceeding even the older, Middle Pleistocene 
forms from the Russian Plain and the Urals (Table 80). This is at variance 
with Bergmann's rule, but the phenomenon may be satisfactorily explained 
402 by the large size of the local prey; Caucasian ungulates. 



ё 10 
К О 

w 

5 




20 31 32 33 34 35 36 37 38 39 40 4f 42 43 mm 

FIGURE 168. Stratigraphic and geographic variation in length of M^of 
Pleistocene and Recent bears 

1 — Ursus arctos drctos, Karelia; 2 — U.arctos caucasicus, 
Greater Caucasus; 3 — U.arctos meridionalis, Lesser Caucasus, 
Talysh; 4— U.arctos arctos (subfoss.), Voronezh, Holocene; 
5— U. karmalkiensis N. Ver. sp. nov., Tatar A. S. S. R. , Nizhnie Karmalki, 
Upper Pleistocene; 6— U. kamiensis N. Ver. sp. nov. , Tatar A.S.S.R . , 
Mysy, Middle Pleistocene; 7 — U. arctos binagadensis, eastern 
Transcaucasia, Binagady, Dots indicate individual measurements; hatched 
circles, mean values 



The smallest glutton was collected in the Middle Pleistocene of the 
middle Urals. 

There are no data on the evolution of the Caucasian martens of genus 
Martes in the Quaternary. However, the Recent martens are represented 
in the area by distinct forms. The skull of the Caucasian pine marten 
(M. martes lorenzi) is on an average 2.0-2.5 mm longer thar the skull 
of the central Russian marten (M. martes ruthena Ogn. ); jugal width 
is greater in M. martes lorenzi. The Caucasian species is 
distinguishable from the Swedish marten (Martes martes martes L.) 
by a more highly-developed facial region, a relatively small brain case, 
and a narrower postorbital bridge (Kuznetsov, 1941). The coloration of 
the fur has more pronounced differences. 



h 


Recent 
1 




Pleistocene 

4 5 
®® 


6 


и - 


: о 


2 


(S> 


л e. 


-i 1— .1 




-Q :o. 


.7) 




L_I 1 L 


I 1 





400 




FIGURE 169. Variation in brain volume of badgers 
from eastern Transcaucasia in the Middle Pleistocene 

1 — Recent M e les meles minor; 2 — fossil 
M . meles aff. minor, eastern Transcaucasia, 
Binagady. Dots indicate the mean values 



These differences should not 
be considered significant, even 
if it is taken into account that the 
species migrated to the Caucasus 
relatively recently, as the 
differences may be due to specific 
features of the Caucasian habitats. 
The Recent pine marten varies 
little in size and color within its 
distribution area in the Caucasus. 

403 - / X \ Individual and age variability 

are much more pronounced, as 
expressed, for example, in the 
shape and color of the neck spot. 

This variability, based on the 
material of the Rostov fur station, 
is shown in Figure 170. 

The biometric differences 
between the Caucasian populations 
of the older stone marten 
(M. foina ne hr ingi) and the 
West European M. foina foina 
Erxl. are, as expected, insignificant. Nevertheless, the skulls of these 
Caucasian martens are 2-2.5 mm larger than those of the Swiss martens, 
and 5-6 mm larger than the skulls of the Crimean martens (M. martes 
rosanovi Martino). 

The skull of the Caucasian mink (Lutreola lutreola turovi)is 
also considerably larger than the skull of the central Russian species 
(Bobrinskii, Kuznetsov and Kuzyakin, 1944). It is known that the increase 
in size is fairly characteristic of populations in newly established marginal 
sections of an area of distribution. 

Detailed morphological and ecological studies are needed for the 
clarification of the genetic affinities between the small and big Caucasian 
weasels: Mustela nivalis caucasica and M. nivalis dinniki. 

404 These often occur side by side and may therefore be regarded as separate 
species. 

Cats such as the panther, cheetah, European wildcat and lynx are of 
interest from the point of view of their origins and speciation in the 
Caucasus. 

The size and proportions of the teeth and skeletal elements of leopards 
have remained practically the same since the Middle Pleistocene (Table 81; 
Figures 61, 6; 85, l), as have the teeth of cheetahs (Vereshchagin, 1951b). 

The Recent Caucasian panthers are smaller, with thinner fur and brighter 
coloration in Transcaucasia than in Ciscaucasia (Dinnik, 1914a; Satunin, 
1915a). They are larger than the south Asian and African varieties. 

The Recent European wildcat of the Caucasus (Felis silvestris 
caucasicus) is generally larger than the European variety 
(F. s. silvestris): the condylobasal skull length of an adult male of the 
former is 99.1 (89.8-102) mm, and of the latter 86-96 mm. The lengths of 
the upper tooth row are 33.1 (30-34.7) and 28-32 mm respectively. 



401 



(40JJxA3LE 80. Stratigraphic and geographic variation in dimensions (in mm) of teeth and lower jaws of 
gluttons of the Caucasus and Eastern Europe " 



Measurements 



Height of jaw behind M i 
Height of jaw near Ртз 

Length Pm2 

Width Pmj 

Length Ртз 

Width Ртз 

Length Mx 

Width Ml 



Upper 
Pleistocene 



и О 2 



30.3 
23.3 

7.1 

5.0 

13.2 

9.0 

24.0 

10,8 



Middle Pleistocene 



О ■-; 1Л 1Л 



л ■^ £ -о 

■-. о v; п) 

Ju о i.^ о 

P N о < 



22.0 
21.5 

9.0 
12.7 
22.0 



28.0 



9.0 



13.5 



24.0 



о 5J 
Z •£ 



о .- 
„ о 
о» О) 



20.5 



19.5-21.0 

17.8 
17.5-18.0 

7.5 
7.3-7.6 

4.8 
4.5-5.1 

10.5 
10.0-11.0 

6.5 
6.0-7.0 

19.5 
19.0-20.0 

9.0 
8.0-9.5 



Recent 



24.0 



21.0-26.0 

19.3 
18.0-21.0 

7.0 
6.0-8.0 

5.3 
4.5-6.0 

11.0 
11.0-11.1 

7.4 
7.0-8.0 

22.1 
21.0-23.2 

9.8 
9.0-10.2 



Note. Mean values in the numerator, observed ranges in the denominator, 
* Author's unpublished material. 



As a rule, cats inhabiting the reed thickets of the Lower Kuban are more 
uniformly colored than those which live in beech forests. Melanic individuals 
occur infrequently in the big beech forests near Nukha, Zakataly and 
Madagiz (Lesser Caucasus). In order to make sound conclusions on the 
scale of isolation of the Caucasian populations, pelts and skulls from the 
north of Asia Minor, the Balkans and the Carpathians must be studied. 
The Recent Caucasian lynx (Felis lynx orientalis)is usually 
regarded as a subspecies of the European lynx. Its body and skull size 
405 sometimes exceed those of the European form, and its coloration is brighter 
than that of the northern form. Morphological changes of the Caucasian 
lynx include decrease in size and simplification of the form of the internal 
nares in the populations of the Lesser Caucasus. 



402 



(404 




|1||'"*"""Ч|| 




к|.ЦИ Illln. ^., 



V V'C"""' ^«"^T'v Ч .f.-.»^ 



I/ 



ii I 1 I I I I ■ ■ ■ ' 



FIGURE 170. Variation in the size and shape of the throat spot of Caucasian pine martens (1) and stone 
martens (2). Numbers indicate frequency 



TABLE 81. Dimensions of teeth (in mm)of Recent and fossil leopards in the Caucasus 



Locality and age 



Species 



Crown length Pm 



Crown length M i 



Western Caucasus, Recent 



Baksan gorge, Sosruko grotto, Early 
Holocene (Mesolithic) 



Upper Rion, Kudaro cave, Middle 
Pleistocene (Lower Paleolithic) . 



Panthera pardus 



P. aff . pardus . 



P . cf. pardus ... 



25.0 



18.1 



24.5-26.2 



17.8-19.3 



18.8 



26.0 



Note. Means in the numerator, observed ranges in the denominator. 



Order LAGOMORPHA 

According to I. Gromov's measurements (1952), the following 
morphological changes have occurred in the European hare of Transcaucasia 
since the time of deposition of the Binagady bitumens: narrowing of the 



403 



facial region of the skull, development of more angular orbits, decrease 
in the volume of auditory bullae, and narrowing of the ilium. The Recent 
European hares of the Caucasus well illustrate Bergmann's rule in that 
they increase in size from south to north. The European hares of 
Transcaucasia (Lepus europaeus cyrensis) have the following 
characteristics: skull length 93-97 mm, liveweight up to 3.5 kg, pale 
yellowish brown in color. The skull length of the Ciscaucasian Lepus 
europaeus caucasicus varies from 95 to 105 mm, the liveweight 
reaches 4.5 kg, and the color is predominantly brownish-yellowish gray 
(Bobrinskii, Kuznetsov, Kuzyakin, 1944). According to our measurements, 
the condylobasal length of the skull of mature hares (females and males) 
from the Kura Lowland in eastern Transcaucasia is 85.3 (79-94) mm (32 
specimens), whereas in western Ciscaucasia it is 88.4 (83-96) mm 
(25 specimens). The liveweight of mature east Transcaucasian hares 
(females and males) is 3.78 kg (14 specimens), whereas the weight of the 
Ciscaucasian hares is 4.76 kg (9 specimens). The increase in the size of 
the hares is most noticeable in the region of the Khachmas Lowland, on the 
Kusary Plain and in the forests of the Lower Samur. This longitudinal type 
of variation confirnns the southern origin of the species. 



Order RODENTIA 

Only few species of post-Pliocene Caucasian rodents are known, so that 
the phylogenetic lineages of most of them are too short to study their 
stratigraphic variation. However, the wide distribution and numerical 
abundance of many of the species permits easy tracing of their geographic 
variability. 

The geographic (latitudinal) variation of morphological and physiological 
features of the little suslik is reflected in the increase in size and in the 
406 darkening of color, as well as in the shortening of the period of estivation 
from semidesert to steppe habitats. 

According to Ognev's measurements (1947), the skulls of the subspecies 
Citellus pygmaeus kalabuchovi from the Sal'sk steppes are on 
average almost 2 mm larger than the skulls ofC. pygmaeus pallidus 
from the Kuma region semidesert. The variation with altitude is even more 
pronounced: the isolated populations of susliks of Ciscaucasia regularly 
increase in size and darken in color with height (Table 82). 

From the semideserts on the Lower Terek to the Upper Baksan the length 
of the body of the susliks increases on an average by 14 mm, tail length 
by 12 mm, foot length by 5 mm, and condylobasal skull length by 3 mm 
(Figure 171). 

However, the mountain populations of susliks also show considerable 
variation in adaptation to hot and dry glacial valleys (altitude 900-1200 m) 
and mountain slopes and subalpine meadows (altitude 2500-2800 m). As a 
rule, populations inhabiting the bottoms of the gorges are similar to the 
plain dwellers, particularly in color. 

The European suslik (Citellus citellus), according to Vinogradov 
and Gromov (1952), increases in size and in the relative length of its tail 
when the populations are traced from east to west (from the Ukraine to the 
Balkans). The Asia Minor subspecies, however, shows pronounced 



404 



morphologic and physiologic variation according to altitude: the body 
increases in size and weight, the tail becomes shorter, and the fur color 
darker (Table 83). According to Avetisyan (1950), the hibernation of the 
suslik in the Alagez area ends by the middle of March, while estivation 
begins in the second half of July. In the upland steppes, hibernation ends 
only in early April. 



TABLE 82. Geographic variation in size (in mni) of little suslik on the Caucasian Isthmus 





Lower Terek 


Dagestan foothills 


Terek-Sunzha 


El'brus slopes 




С i tellus 


С itellus 


Plateau 


С itellus 


Dimensions 


pygmaeus 


pygmaeus 


С itellus pyg- 


pygmaeus 




planicola 


sa t u n i n i 


maeus boehmii 


musicus 




25 specimens 


40 specimens 


7 specimens 


48 specimens 




206.7 


208.5 


207.5 


220.0 


Body length 


182-230 


182-230 


193-215 


205-240 




35.6 


36.3 


37.8 
28-40 


47.4 


Tail length 


30-40 


32-43 


45-50 


Foot length 


32.1 


32.1 


33.2 


37.4 


27-35 


30-36 


28-34 


36-38 


Condylobasal skull length 


40.7 


40.0 


41.0 


43.5 


39.5-42.0 


37.9-42.6 


37.8-41.9 


42.5-45.2 



Note. Means in numerator, observed ranges in denominator. 



Some general evolutionary trends and specialization rates can be studied 
in the jerboas (genus Allactaga). 



407 



TABLE 83. Changes in size (mm) and weight (g) of Asia Minor suslik (males and females) at different 
altitudes 



Dimension 


Zone 1255-1550 m 
above sea level 


Zone 1550-2190 m 
above sea level 




203 


215 


Tail length 


175-230 
45 


180-390 
43 




33-37 
239 


21-59 
278 




188-372 


184-430 



Note. Means in the numerator, observed ranges in the denominator. 



405 




FIGURE 171. Variation in size of little suslik in Ciscaucasia. Rectangles indicate the length of a foot 
squared. Populations from the Lower Terek taken as an entity. Black dots — points of trapping 

The jumping ability of Recent jerboas of Ciscaucasia and the Apsheron 
Peninsula has improved since the Middle Pleistocene (Binagady) due to the 
increased length of the tibia relative to the femur, as shown in the increase 
408 in the jumping index (length of tibia/length of femurX 100) of the small and 
Asia Minor jerboas (Table 84). 

I. Gromov (1952) has pointed out that the structure of the metapodia in 
the great jerboa from the Binagady asphalt (Allactaga jaculus 
bogatschevi) is more primitive than in the Recent form in the wider 
spacing of the distal part of the foot and in the longer free end of the second 
metatarsal (mtt2). 

Hamsters (Cricetinae) have occurred in the Caucasus since the Middle 
Miocene. It is difficult, however, to establish the affinity of the Miocene 
(Palaeocricetus, Belomechetka) and Pliocene (Stavropol) hamsters 
with the Quaternary genera Cricetus and Mesocricetus. Hamsters 
from the Lower Pliocene alluvium of Stavropol (Kosyakin quarry) are 
characterized by large molars. The hamster from the Binagady asphalt 
(Mesocricetus auratus planicola) is intermediate in size between 
M. auratus raddei and M. auratus nigriculus. The Binagady 
species differs from the forms of Mesocricetus known in the U. S. S. R. 
in its hard palate and short incisor foramina. The structure of the 
postcranial skeleton of the Upper Pleistocene hamsters from the Chiaturi 
manganese region indicates that the population was no less specialized than 
the Binagady population. 



406 



TABLE 84. Changes in jumping index of Middle Pleistocene jerboas* 



Species 


Jumping index 


Binagady 


Recent 




125.0 
126.1 
129.5 




127.0 


A.williamsi 


129.5 


A.elater 


136.8 







We calculated the indexes from the mean values given by I. Gromov (1952); he tried to reconstruct 
the ancient landscapes of the Apsheron on the basis of certain morphological features of the Binagady 
jerboas. His assumptions, however, on the existence of either soft ground, mosaic landscapes or 
diversified relief at Binagady are poorly founded . 



409 



The geographic variation in the Recent populations of hamsters in Asia 
Minor is mainly reflected in changes in the depth and brightness of the 
rusty color of the back and sides, and the black color on the abdomen 
(Argiropulo, 1937). Also the mountain populations are, as a rule, larger 
than those of the Ciscaucasian foothills, though the ecological variability 
of the species is great. On the high Armenian plateaus, hamsters begin 
to hibernate at the end of October or early November, whereas on the lower 
Kakhetian plateau they begin only in early December. 

The Recent gray hamsters of the Apsheron Peninsula have developed 
a narrower anterior palatal region, larger molars, a more protruding 
coronoid process, and a slightly recurved lower incisor and M3 with 
shortened heel. These changes have occurred since the Middle Pleistocene. 
The functional significance of these minor changes and the evolutionary 
trends of the species are difficult to explain. It is simpler to record the 
facts, as has been done by I. Gromov (l952). 

The process of morphological and ecological evolution of the species as 
polymorphic as the house mouse (M u s mus cuius) in the Caucasus is 
very complex. This species was widely distributed in the Middle Pleistocene 
of eastern Transcaucasia (Binagady), and also occurs in this region today. 
The morphological and ecological differentiation of the species has led to 
evolution of the following subspecies: white abdomen and short-tailed 
(steppe and semidesert forms), M. musculus hortulanus, 
M. musculus tataricus; gray abdomen and long-tailed (mountain- 
forest forms), M. musculus abbotti, M. musculus formosovi; 
and introduced domestic forms, M. musculus musculus. This 
variability is as developed as the variability of the common field mice of 
the Caucasus (discussed below); however, it cannot yet be studied using 
the modern methods of morphological investigation (see Geptner, 1930; 
Gulii, 1930; Sviridenko, 1935a). 

The systematics of the common field mice of the Caucasus (genus 
Apodemus) has not been extensively studied, but on the basis of the 
studies of Argiropulo (1940a, 1946), Sviridenko (1936), Kuznetsov 
(Bobrinskii, Kuznetsov and Kuzyakin, 1944), Shidlovskii (1953a) and our 
observations, the genus may be considered as being subdivided into four 



407 



species: A. mystacinus, A. flavicollis, A. fulvipectus and 
A. sylvaticus (see also Chapter V). This classification is tentative 
and subject to future experimental ecologic studies. 

The Asia Minor mouse (A. mystanicus), distributed in the Balkans, 
Asia Minor and western Transcaucasia, is considerably smaller in size 
in the Caucasus than in the Balkans. According to Shidlovskii's 
measurements (1953a), the condylobasal length of the skull of the Balkan 
specimens is 28.4 mm (34 specimens), and of the Transcaucasian specimens 
26.5 mm (200 specimens). The length of the upper molar row is 4.9 and 
4.5 mm respectively. The Balkan populations are lighter in color than 
those in Asia Minor and Transcaucasia. According to Shidlovskii (1953a), 
this animal migrated to the southern slopes of the Greater Caucasus 
(Map 55) recently, during the last few decades. In the present case, the 
body size of the animal, which migrated into a new mesophytic environment, 
has decreased considerably. Paleontological studies of the Imeretian caves 
will clarify whether this is a primary or a secondary immigration and will 
consequently shed light on the degree of morphological divergence of this 
east Mediterranean species. 

The picture of the geographic variation of the other three species on the 
Caucasus Isthmus is much more complex. The yellow -necked mouse of the 
European type, with feet 24-25 mm long (close to A. flavicollis 
samariensis) occurs only in the river valley forests on the Lower Sulak 
in eastern Ciscaucasia. * In the far southwest the southern boundary of the 
distribution of this species reaches the Lower Don (Map 55). 

The relatively small A. f u 1 v i p e с t u s*='S with a foot 22-23 mm long, 
occurs in the deciduous forests of the lower mountain belt in the Krasnodar 
area, in Abkhazia, Adzharia and northwestern Azerbaidzhan, and in some places 
in northern Armenia (Map 56). This species is easily distinguishable from 
the common field mouse, even in juvenile states, by its larger dimensions, 
greater length (relatively shortened) of the foot and the presence of an oval 
yellow spot on the breast. The species comes out into the sun to feed only 
rarely. It is a good climber and often lives in hollows in trees. It also 
differs from the common field mouse in its weaker response to light. 
'*1^^ It seems likely that the relatively short foot and tibia, the longer tail of 
the yellow -spotted mouse and its relatively narrow internal nares 
(Argiropulo, 1946), as compared with these features of the common field 
mouse, are adaptations to slower movement and better climbing. 

The populations of this species vary even within their area of distribution. 
As a rule, the southern forms are somewhat smaller, while to the east 
their size increases. Morgilevskaya (1954)has published data on five mixed 
populations of the yellow -spotted and common field mice. According to 
her data, the condylobasal length of the skull is (accordingly to locality): 
Sukhumi — 23.8, Gagry — 22.3, Batumi— 23.1, Akhalkalaki — 22.7, 
Lagodekhi — 23.7 mm; length of hind foot: Sukhumi — 22.6; Gagry — 21.3, 
Batumi — 21.4, Lagodekhi — 22.7 mm. 



This is trae only if the labels on the material in the Moscow Zoological Museum have not been mixed. 
Geptner and Formosov (1941) have identified 5 specimens from this region and I studied one specimen, 
but the origin of this species is not yet clear to me. 

We have assigned to this species the Caucasian subspecies A. flavicollis ponticus, A. flavicollis 
parvus and A. sylvaticus fulvipectus of earlier authors. 



408 




Populations of ye How -spotted 
mice are common in a number of 
valleys in Central Ciscaucasia 
(Urukh, Ardon, Terek) and on the 
piedmont plains of eastern 
Ciscaucasia. In this region the 
animals vary considerably in size 
from one valley to another, which 
is due to both genetic isolation and 
differences in habitats. The higher 
degree of ionization of the air in 
the area of tectonic fracture has 
been suggested as the reason for 
the larger size of the mice in the 
Armkha ravine (tributary of the 
Terek) (Ushatinskaya-Dekalenko, 
1933). If this is true, then the size 
of mice in the mountain populations 
may be an indicator in surveys of 
radio-isotopes . 

In eastern Transcaucasia this 
species primarily occurs in the 
piedmont lowlands. Careful 
measurements in a series of 
preserved specimens of 
A. fulvipectus and common 
field mice of the same age group 
gave the data on the geographic 
variation in foot length, as shown 
in Figure 172. 

Isolated bones of the common 
field mouse (A. sylvaticus 
subsp. ) occur in the Middle 
Pleistocene asphalts of the Apsheron 
Peninsula. No morphological 
differences between these and the 
Recent forms have been recorded 
in this material. In Ciscaucasia, 
this species is widely distributed 
on the plains, in the foothills and 
in the mountains, particularly in 
areas of sparse vegetation (Map 57). 

The mountain populations of this species are characterized by higher counts 
411 of erythrocytes in their blood (Kalabukhov, 1940). In Transcaucasia the 

common field mouse is somewhat larger, which sometimes makes it 

difficult to distinguish from A. fulvipectus. A yellow spot is almost 

always present on the breast of mature specimens. 

The Caucasus is thus an ancient and self-contained center of speciation 

of domestic and field mice; this confirms their Pliocene origin in this 

country. 



. 




IQy 


У ^ : 

• ••III 



18 19 20 Z1 Z2 23 24 2S 26 



FIGURE 172. Variation in foot length of common field 
mice on the Caucasian Isthmus 

Apodemus flavicollis: 1 — Rostov region. 
A. fulvipectus: 2— Caucasian Reservation; 

3 — Alazan-Agrichai valley (Transcaucasia); 

4 — Lesser Caucasus (Delizhan); 5 — Talysh. 

A. sylvaticus: 6 — Rostov region; 7 — Caucasian 
Reservation; 8 - North Ossetia (Lars, Zamarag); 
9 — Alazan-Agrichai valley (Zakataly); 10 — middle 
Araks valley . On the left — foot of A. sylvaticus; 
on the right — A . fulvipectus. Ordinate — 
number of specimens 



409 



The red-tailed gerbil from the Binagady bitumens (Meriones 
erythrourus intermedius) is a form intermediate between the Recent 
gerbils occurring to the west and east of the Caspian. 

The primitive features of the Middle Pleistocene gerbil are reflected 
in the shortness of the tibia, which is only 113.8% the length of the femur, 
as compared with 125.9% in the Recent gerbils of the Apsheron. * 

Insofar as the lengthening of the tibia relative to the femur indicates 
an adaptation for jumping (jerboas), it can be stated that the Recent gerbils 
became "jerboaized" (N.A. Smirnov's term) since the Middle Pleistocene. =■'* 

Evolution in space and time can be studied in the case of the Caucasian 
voles (genera Pr ome the omys , Arvicola, M i с r о t u s) as rewardingly 
as in mice. 

The teeth and lower jaws of Prometheomys from the Acheulean beds 
of Kudaro I and the Upper Paleolithic beds of the Gvardzhilas cave are not 
noticeably different from the Recent forms. This is understandable, since 
the morphological features of burrowing rodents reflect conservativeness 
in habit. The data on the geographic variation in the color of Recent 
populations are summarized in Table 85. 



TABLE 85. Variation in abundance of melanic varieties in the distribution area of Pronnetheomys * 





Upper Laba and Belaya 


Upper Bzyb 


Upper Terek 


Color 


no. 


iTj 


no. 


% 


no. 


1o 


Grayish brown 

Black 


13 

12 


52 
48 


16 


100 


15 


100 



* Collections of ZIN AN SSSR. 



412 



The increased proportion of the melanic varieties is probably restricted 
to the western part of the distribution range. 

The Recent populations of water vole (Arvicola t e r r e s t г i s) vary 
noticeably within the Caucasian and adjacent regions. At least 9 varieties 
have been described in the Caucasus, and not less than 6 varieties from 
neighboring areas in Iran, Turkey and the Russian Plain. It is difficult, 
however, to trace the geographic variation in the morphological characters 
of the species. 

As a rule the northern marshland forms from the Lower Don, Kuban 
and Volga are somewhat larger than the mountain and south Caucasian 
forms (Table 86). 

Populations of this species from some sections of the distribution range 
are characterized by minor morphological adaptive features. The upper 
incisors of the mountain "burrowing" populations of Balkaria, Ossetia, 
and Dagestan resemble those of mole voles. The curvature of the incisors 
is greater than in the marshland forms, and the degree of differentiation 



Indices of the mean values were computed by us from the measurements given by I. Gromov (1952). 

I. Gromov has also indicated that the Binagady gerbil was of a "more running" type (i.e., less of a jumping 

type) than the Recent species. 



410 



of the fur is less marked. The change in shape of the incisors is due to 
their adaptation for digging. 



TABLE 86. Variation in skull size of the water vole in the Caucasian Isthmus* 



Dimensions 


Lower Don 

Arvicola 

terrestris 

tanai t icus 

10 specimens 


Lower Kuban 
Arvicola 
terrestris 
cubanensis 
11 specimens 


Eastern Dagestan 
Arvicola 

terrestris 
kurusch i 

32 specimens 


Transcaucasia 

Arvicola 

terrestris 

persic us 

32 specimens 




41.2 


41.0 


38.5 


39.6 


Condylobasal length of skull . . . 
Length of upper molar row . . . 


41.0-41.5 

10.2 
10-10.4 


39.5-42.6 
9.9 


38.2-38.3 
9.4 


37.3-41.9 
10,1 


9.2-10.5 


9,6-10.8 



Note. Means in the numerator, observed ranges in the denominator, 

* Measurements by Ognev (1933). 



Up to 8 subspecies of the snow vole known from Palestine to Kopet-Dagh 
are also found in southwest Asia, the Caucasus and Central Asia, The 
geographic variation of the Asia Minor snow vole occurring between the 
Lesser Caucasus and Kopet-Dagh is expressed in lighter coloration correlated 
with increasing draught and isolation, in the shortening of M^ along the 
west — east gradient, and in a general decrease in size (Table 87). 



(413) 



TABLE 87. Geographic variation in size (in mm) of Asia Minor snow vole* 



Dimensions 


Northwestern 

Caucasus, 

Mt. Pambak 

M icrotus nivalis 

loginovi 

7 specimens 


Lesser Caucasus, 
Lake Sevan shores 
M icrotus 
nivalis 
satunin i 
8 specimens 


Talysh, 
Mt. Kelakhan 
M icrotus 
nivalis subsp . 
1 specimen 


Kopet-Dagh, 
Mt. Shakh-Shakh 
M icrotus 
nivalis 
dementie vi 
3 specimens 


Body length 


109.4 
99-121 

58.1 
52-68 

18.8 


121.9 
109-142 

54.0 


115J3 

52J3 

18.5 

27.3 

6.3 


103.2 




Tall length 

Foot length 

Total skull length 

Length of upper molar row 


97- 120 
55.3 


32-90 
17.9 


51-60 
18.5 


12.2-20,0 
26.4 


13-20 
27.8 


17-20 

27.2 


25.0-27.8 
5.9 


26.3-29.6 
6.3 


25.8-28.5 
6.1 


5.5-6.2 


6.1-6.7 


5.5-6.5 



Note. Means in the numerator, observed ranges in the denominator. 
* Author's measurements on the ZIN material. 



411 



413 



The snow vole from the Greater Caucasus is the smallest variety, 
indicating that humid mesophytic regions are not suitable for this animal. 
The decrease in size of this species towards the southeast, and its 
disappearance from the intermediate heights of the southern slopes of the 
Arn^enian uplands, indicate that the southeastern sections of the present 
distribution area are in a state of decline. This also indicates that in the 
past a more humid phase existed, during which the distribution area of the 
species extended further to the east. 

The geographic variation in the Caucasian snow vole on the Greater 
Caucasus is reflected in the decrease in size and lighter coloration from 
west to east. The change in color is probably due to the fur becoming 
bleached in sunny woodless Dagestan. The inversion of the zones in the 
central part of the Greater Caucasus is somewhat superimposed on the trend 
of brightening of the color. 

The decrease in size from west to east is shown by the measurements 
given in Table 88. 

It should, however, be mentioned that size variation of this species of 
vole has not yet been traced through the altitudinal and landscape zones. 



TABLE 88. Geographic variation in size (in mm) of Caucasian snow vole* 



Dimensions 


Western Caucasus 

M icrotus gud 

nen jukovi 

128 specimens 


Central Caucasus 
Microtus gud 
gud 
116 specimens 


Eastern Caucasus, 
Dagestan, 

Microtus gud 
Ighesicus 
8 specimens 


Body length 

Tail length 

Foot length 

Condylobasal length of 

skull 

Length of upper molar 


120-152 

68-106 

19.0-26.9 

26.4-28.6 

6.7-7.8 


102-149 
61.0-85.0 
18.0-23.0 

25.0-29.2 

6.0-7.0 


100-121 

58-78 

18.5-21.2 

26.4-27.2 

6.0-6.6 







Measurements of Ognev (1950) and author's (for Dagestan). 



The Middle Pleistocene Apsheron vole (Microtus (Pitymys) 
apscheronicus) differs from the Recent pine vole of the Caucasus in its 
larger size, its longer tooth row (6.3 mm as compared with 5.7 mm of the 
Recent species) and in details of the shape of the lower jaw. It is also 
possible that the extinct species was of a less mesophilous character than 
the Recent form. 

Geographic variation of the Recent pine vole is poorly developed. The 
subspecies described for the Caucasus reflect the subjective approaches 
414 of investigators or the ecological variants (and age groups) arising due to 
variation in living conditions from year to year. The subspecies described 
are: M. majori majori from northeast Asia Minor, the Kars Highland, 
Dzhavakhetia, Mingrelia, and the central and western parts of the Greater 
Caucasus; M. majori ciscaucasicus from the northern slope of the 



412 



Bol'shoi Range, in its western and central parts; M. majori suramensis 
from the central sections of both slopes of the Bol'shoi Range, Surami 
and the Lesser Caucasus; M. majori dagestanicus from inner 
Dagestan, Tushetia, northwestern Azerbaidzhan, the northeastern slopes 
of the Lesser Caucasus and Dhavakhetia; M. majori schelkovnikovi 
from the forests of Talysh. 



TABLE 89. Geographic variation in size (in mm) of 


common vole in 


the Caucasus* 




Dimensions 


Foothills of nonhern 

Caucasus 

M icrotus arvalis 

macrocraninus 


с 

6 

о 
a) 
a, 

(M 

CO 


> 

с 
с 


central Caucasus 

M icrotus arvalis 

gudauricus 

52 specimens 


Armenian Highland 
M icrotus arvalis 
transcaucasicus 
133 specimens 


Talysh Highland 
Micro tus arvalis 
mystacinus 
4 specimens 


Condylobasal length of 
skull 


26.3 






23.8-27.3 
5.7 


25.3 


25.2 


25.8-27.3 
6.0 


24.0-26.9 
5.9 


24.0-26.0 


Length of upper molar 


5.5 


5.8-6.5 
17-18.2 


5.3-6.2 
15.5-19.6 


5.2-6.0 
14.1-18.0 


5.2-5.7 




16.2 


Foot length 


15.0-17,0 



Note. Means shown in the numerator, observed ranges in the denominator. 
* Measurements of Krasovskii (1930), Ognev ( 1950) and author's (for Talysh). 



415 



The Talysh specimens are distinguishable by their intense rusty color 
on the sides and their small size. M. majori dorothea Ellerm. from 
the Elburz Range is probably synonymous with the latter subspecies. The 
pronounced differences between the M. schelkovnikovi subspecies 
and the pine voles of the Lesser and Greater Caucasus attest to the long 
period of isolation of this animal in the southeastern section of its 
distribution range. The overlap of the distribution ranges of the former 
four subspecies is self-explanatory. The rather vague systematic diagnosis 
of the forms described indicates that in order to understand their geographic 
variation special ecologic and morphologic studies must be made. 

It is known that the common vole (M i с r о t u s ar v al i s) decreases in 
size from north to south on the Russian Plain. On the Caucasian Isthmus, 
with its diversified relief and inversion of the zones, Bergmann's rule does 
not strictly apply to the distribution of the species. Studies of the geographic 
variation of this species must be done systematically across the altitudinal 
and landscape zones, taking into account seasonal and annual variation. 

Some numerical results of the studies of the Caucasian subspecies are 
given in Table 89. 

Animals inhabiting the dry longitudinal valleys east of the El'brus and 
inner Dagestan are characterized by a lighter color than those inhabiting 
the humid foothills. 



413 



The type of geographic variation described and the characteristic features 
of the area of distribution attest to the ancient origin of the species on the 
mesophytic uplands of the Mediterranean. 

The Middle Pleistocene steppe vole, Microtus socialis, from the 
Binagady asphalt is almost identical in skull structure to the Recent species 
(Gromov, 1952). 

According to Ognev (1950, p. 400) these subspecies increase in size from 
north to south; their auditory bullae also increase in size and the structure 
of M^ and M^ becomes more complex. However, these facts need further 
study, particularly in relation to the specific ecologic features and to height 
above sea level. Considering the great size of the total area of distribution, 
the variation mentioned is not particularly significant. 

All this indicates that the relationships between the steppe vole and the 
xerophytic landscapes of the Mediterranean have remained stable; they 
date at least from the Lower Pleistocene. 



Order PROBOSCIDEA 

In the south of the Russian Plain, in the Caucasus and in southwest Asia 
(as in the Mediterranean in general), the first known true elephants are 
represented by highly specialized forms: Elephas (Archidiskodon) 
planifrons, E. (H e s pe r о 1 о x о d о n) an t i q u u s , E. (Archidiskodon) 
meridionalis from the Middle Pliocene (Trouessart, 1898- 1899c; 
Pavlova, 1910a; Bogachev, 1923-1924; Osborn, 1942; Simpson, 1945). 
The small elephant (Phanagoroloxodon mammontoides), as 
described by Garutt (1957a) from some Upper Tertiary beds of western 
Ciscaucasia^'-, is probably of the same age. The affinities of this species 
to those mentioned above are not clear. However, the species is considered 
to be close to E. (Hesperoloxodon) antiquus or E. (Parelephas) 
t r о go nt he r i i , so that Garutt's identification of this form as Early 
Pleistocene is undoubtedly false. The absence of direct ancestors of 
the species listed in the Oligocene, Miocene and Lower Pliocene beds 
is probably not due simply to change. It is worth noting that the 
remains of elephants definitely do not occur in the immediately post-Pontian 
beds in Ciscaucasia and in the northern Black Sea area as they do, in the 
alluvium of the Kosyakin quarry near Stavropol and in the clay fissure - 
fillings in the Pontian limestone, exposed in the Odessa catacombs. In spite 
of the "taphonomic universality" of both burials, the proboscideans are 
represented only by mastodons (and Dinotherium at the Kosyakin 
quarry). 

Evolution of the elephants in the region must be studied mainly from 
tooth material, on which the poorly-developed hypothesis of the evolutionary 
lineage E. planifrons-E. meridionalis-E. trogontherii- 
416 mammoth has been based. The systematic position of the so-called ancient 

" The Phanagorian elephant has been described fium a permineralized skull (with broken teeth) of 
unknown locality in the Krasnodar museum. Garutt's (1957a) photographs show teeth of elephants 
collected from the conglomerates near Sennaya and Akhtanizovskaya on the Taman Peninsula, but his 
identification of the material as belonging to the Phanagorian elephant is not certain; these teeth seem 
rather to belong to the southern and ancient elephants (Vereshchagin, 1957a). There is unfortunately 
no space for discussion of the opinions of Sherstyukov (1954), which are lacking in clarity. 

414 



elephant [E. antiquus] is not sufficiently clear. Russian paleontologists 
have either identified this form with the Loxodonta group or have related 
it to E. planifrons and E. meridionalis. It has been identified 
as P a 1 a e о lo X о d о n by Dubrovo (1957). Because of the scarcity of 
remains of the ancient elephant, it is only possible to construct a 
distribution map (Map 76). Over 400 casts and descriptions of various teeth 
of Upper Pliocene and Quaternary elephants, collected from the Russian 
Plain and the Caucasus, have been studied by us in the Russian museums 
and collections. In addition to distribution maps showing occurrence of 
teeth (Maps 74-76), we can give a general description of the teeth of various 
phylogenetic grades and a summary in table form of the main morphological 
features. Most teeth and bones of Pliocene and Quaternary elephants in the 
Caucasus occur in redeposited sediments. АЦ the material has been washed 
out of the primary bone -bearing lenses by streams and redistributed in 
conglomerates, gravels and sands. The localities with E. planifrons 
and E. meridionalis in Ciscaucasia are situated in delta zones of 
ancient streams and mudflows, which once flowed from the Greater 
Caucasus to the Akchagyl and Apsheron seacoasts. The bones often occur 
in ferruginous gravels and sands of the high (third or fourth) river terraces 
on the inclined piedmont plains. According to Bogachev (1923-1924), most 
of the bones of southern elephants on the coast of the Sea of Azov (Lower 
Don) have been washed out of the Middle Pliocene beds and redeposited in 
Late Pliocene gravels. 

In eastern Transcaucasia the bones occur mostly in loesses, loams 
and gravels, around the ancient coasts of the Kura bay of the Akchagyl and 
Apsheron seas. On the Lesser Caucasus uplands, bones of Upper Pliocene 
and Lower Quaternary elephants occur in river and lake-river sediments, 
like those near Leninakan and Erzurum. Elephant bones occur at Tsalka 
in conglomerates between doleritic lava flows, dated as Giinz and Gunz- 
Mindel (see also Burchak-Abramovich, 1951a). 

Teeth of Elephas planifrons ''= and E. meridionalis are 
relatively small, with short and weak roots. Their enamel pockets are 
coarse -ribbed, comb-like plates, formed of enamel cylinders intergrown 
longitudinally. With a covering of dentine and cement, these widely-spaced 
pockets form enaraiel bellows; upon wearing, the pockets became 
transversally flattened, forming enamel loops with thick walls protruding 
5-6 mm above the surface in mature individuals (Figure 173). The working 
surface of the tooth was used for grinding large pieces of plant material. 
The bases of the enamel pockets, near the pulp cavity, form a uniform 
roller with poorly-developed papilla. The cement of the teeth of these 
early elephants has been altered and weakened to such an extent that the 
roots, as a rule, have been destroyed and the pulp cavity is full of earth. ** 
On the Russian Plain remains of Upper Pliocene elephants occur mostly in 
the south, none having been recorded north of the latitude of Moscow and 
Kazan. 
417 As far as diet is concerned, E. planifrons and E. meridionalis 

were probably still fairly close to the mastodon. Striations and scratches 
in the enamel of the teeth of these elephants strongly resemble those on 
mastodon teeth, though in mastodons the grooves are usually transverse 

" Garutt (1957b) has expressed an opinion, not yet well founded, that the so-called E. planifrons 
known fronn the U.S.S.R . is merely an early form of E. meridionalis, 
** The cement and dentine of these teeth usually resemble alabaster and marl. 



415 



to the jaw, while in elephants they are longitudinal. This is probably due 
to differences in the mode of grazing. The grooves were undoubtedly formed 
by sand grains and small stones taken up together with roots and grasses. 

The seasonal feeding of elephants in the thickets and swamps and along 
the shores of lakes and in river valleys undoubtedly explains their relatively 
frequent burials in water-laid sediments. 

Tooth replacement in Upper Pleistocene elephants has not yet been studied 
thoroughly, but it is generally assumed that in each jaw 5 molars were 
worn off and replaced by the appearance of new molars from behind and 
above in each jaw, in a manner similar to the replacement of the molars 
of the Recent Indian and African elephants. The last, 6th tooth, the biggest, 
with thickest enamel, was characterized by the absence of a posterior 
crushing surface; it functioned only in the last stages of the animal's life. 
This type of molar replacement, which was probably also characteristic 
of mammoths, is taken into account in the discussion which follows. * 

Towards the Lower Anthropogene, the number and size of enamel pockets 
and consequently the height of the teeth have considerably increased; the 
enamel wall has become thinner and less folded. The teeth, particularly 
the lower teeth, have become relatively narrower and longer, and their 
volume and weight have increased both absolutely and relatively. This stage 
in the morphogenesis of elephants is represented by a series of burials on 
the lower (second and third) river terraces ofElephas (Parelephas) 
w u s t i and E. (Parelephas) trogontherii. *" The occurrences of 
bones of E. trogontherii are best known in the gravels of the third 
418 terrace on the left bank of the Dniester, near Tiraspol, and in the Araks 
valley, near Yerevan and Leninakan (Map 75). 




FIGURE 173. Section through M of the southern elephant from the northern coast 
of the Sea of Azov 

Section shows enamel pockets and specific features of loop formation (a — loop; 
b — pocket) 



Fragment of the skull of a southern elephant from Khapry (collections of Institute of Geological Sciences 
AN SSSR ) shows that the replacement of the second upper tooth by the third took place directly from 
above, similarly to the mode of replacement in mastodons, rather from than above and behind as in 
mammoths and Recent elephants. 

The latter name poorly reflects the actual situation, since the increase in abundance and area of 
distribution of the "trogontherial" beaver occurred probably as early as the Upper Pliocene. 



416 



The Middle Pleistocene sees the beginning of the final stage in the 
evolution of elephants. It coincided with the rapid development and 
proliferation of Middle Paleolithic man, who rapidly perfected the art of 
hunting, including the elephant hunt. By that time E. trogontherii had 
decreased in size and evolved into the mammoth, E. (Mammonteus) 
primigenius, and had completely changed in appearance, having 
developed a woolly coat='% longer and more sharply curved tusks, and 
stouter m.etacarpals and metatarsals. The evolutionary trends in the teeth 
included closer spacing of the enamel loops and increase in their 
number per unit length, thinning of the enamel, and smoothing of 
the enamel surface (Figure 174). Sharp edges developed on the bottom 
of the enamel pockets; pointed papillae situated along the sharp rim 
protruded into the pulp cavity. The teeth roots developed into long, 
curved or straight hollow tubes and the closely spaced enamel 
lamellae make a finer, and possibly more durable, grinder. Traces of 
vegetable matter are found only rarely on the truncated surfaces of the thin 
enamel of the teeth of mammoths and E. trogontherii. While feeding 
on twigs and tree branches undoubtedly continued, grass as a diet began 
to gain in importance, as has been definitely established by the contents of 
the stomachs of the frozen mammoths in Siberia. 

Remains of true mammoths in the Caucasian Isthmus are more common 
in Ciscaucasia than in Transcaucasia. As as rule, they occur in the young 
(first and second) river terraces. The state of preservation of the enamel, 
dentine and cement of the mammoth teeth is always strikingly different 
from the preservation of these components in the teeth of earlier elephants, 
as roots and the pulp cavities of mammoth teeth are often preserved in their 
entirety. 

The earliest remains of mammoth-type elephants from the Caucasus 
were found in the Girei quarry on the second terrace of the Kuban. Only 
the materials from the Il'skaya site in the Trans -Kuban region have been 
dated. Four fragments of molars, many pieces of tusk, and up to 40 
assorted incomplete skeletal elements including tarsals and carpals have 
been collected from this site. 

The last occurrence, stratigraphically speaking, is part of a left upper 
jaw with a very worn 6th tooth fronn the travertines of the northern slope 
of Mt. Mashuk. The whitish enamel, the light cement— light brown 
color of the bone in fresh fracture, and the characteristic smell of raw 
bones upon scratching — all these features are highly unusual for a burial 
so far south. The small size of the tooth, the thin enamel (1.0-1.2 mm), and 
the small diameter of the tusk indicate that the fragment belongs to a 
stunted, very late variety of mammoth of Late Paleolithic and even Neolithic 
age. 

In Transcaucasia, mammoth teeth have so far been found only in the Gori 
depression (Burchak- Abramovich, 1946; Gabuniya, 1952a). The tooth 
described by Gabuniya (1952a), an unworn M'^, was collected on the first 
terrace of the Kura River in the town of Gori. On the basis of the small 
size of the tooth with 12 lamellae over 10 cm and enamel thickness of 1.5 mm 
Gabuniya has related the tooth to the "stunted" mammoths of the west 
il9 Mediterranean. On the basis of the mode of its occurrence the specimen 
has been dated as early post-Wiirm, i.e. , Lower Mesolithic to Neolithic. 

* So far there is no information on the woolly coat of E. trogontherii. 



417 



Gromova (1948) has assumed, without any evidence, that "some mammoths 
lived in the Caucasus in pre-Riss time. " Burchak-Abramovich (1946) 
considered (incorrectly) that the Transcaucasian mammoth is a 
representative of the northern tundra. At variance with these investigators, 
Gabuniya has correctly emphasized the probability of evolution of small 
local forms of mammoth in the south of eastern Europe and in Transcaucasia. 




I 1 I I ' I ' I ' ■ ' ^'"1 

FIGURE 174. Longitudinal sections 

1 — M; 2 — Mgof marnmoths from the Russian Plain 



418 



420 



Even if the northern mammoths migrated to the Caucasus during the 
maximum height of the cold periods, it is doubtful that they could migrate 
farther south than Ciscaucasia. In the Upper Pleistocene the western 
migratory route of the northern mammoths through forests and steep rocks 
was difficult, and the eastern route passed through a country too dry and 
hot for mammoths. 

Thus Transcaucasia was probably the southern boundary of the 
distribution of the mammoths, and this is confirmed by the absence of 
reliably identified remains of mammoths from the Armenian Highlands, 
Iran and Southwest Asia in general. The xeromorphic environment with 
poorly-developed wood vegetation, which evolved (at least in the low valleys 
of these regions) towards Middle Pleistocene time, was not suitable for 
the elephants of late and northern types. 

North of the Ciscaucasian plain, mammoth remains are the most common 
paleontological finds in the valleys of the Don, the Volga and the Ural, and 
farther north the proportion of mammoth bones also steadily increases by 
comparison with the bones of earlier elephants. The results of our counts, 
done in 1950-1955, are given in Table 90. 



TABLE 90. Number of teeth of fossil elephants in regional museums 



Museums 



Mammoths 



E 1 ep ha s 
trogontherii 



E lepha s 
meridionalis 



Arkhangel'sk 

Veliki Ustyug 

Vologda 

Kazan University 

Ural'sk 

Kuibyshev 

Saratov 

Astrakhan 

Novocherkassk 

Temryuk and collections from 
Taman and at the Paleontological 
Institute 

Krasnodar 

Stravropol 

Pyatigorsk 

Makhachkala 



12* 
33' 
23» 
59 
10 
10 
21 
51* 
3 



100.0 
100.0 
100.0 
98.4 
83.3 
77.0 
81.0 
43.5 
18.7 



7.3 
33.3 
25.0 
44.5 
33.3 



1 


1.6 


2 


16.7 


3 


23.0 


5 


19j3 


71» 


56.5 


4 


25.0 


8 


19.5 


11 


52.4 


2 


16.7 


2 


22.2 



30 
3 
7 
3 
2 



56.3 



73.2 
14.3 
58.3 
33.3 
66.7 



Note, Asterisk indicates cases when bones other than teeth were counted. 



The distribution of the species, as given in Table 90, indicates more 
than the apparent decrease in the frequency of burial of mammoths in the 
south and the increase in erosion and exposure of older beds containing 
remains of southern elephants. Recorded distribution of the proportions 
of the species is due to the fact that the mammoths were very abundant in 
the north, the main area of their distribution, whereas the distribution areas 
of E. trogontherii and E. meridionalis were further south. 



419 



(421) 




FIGURE 175. Print of worn surface (enamel loops) and outlines of teeth Mg of fossil elephants 
of the Russian Plain and the Caucasus (M^— a; Mg— b) 

Elephas (M am mon teus) pr im igen ius: la, lb — Sukhona valley, No. 10401, 10870, 
Tot'ma Museum; 2a — peat in Don valley, un-numbered, Voronezh Museum; 3a — Kirillovskaya 
Upper Paleolithic site near Kiev. No. 302, Institute of Zoology of the Academy of Sciences 
of the Ukr. S. S. R. 



420 



(422) 




FIGURE 175 (continued) 

4b — Middle Paleolithic site, Chokurcha, in the Crimea, un-numbered, Simferopol Museum; 
5a — second terrace of Kuban, Girei quarry, near Kavkazskaya, un-numbered, Armavir 
Museum; 6a — Terek valley near Naurskaya, un-numbered, Georgian Museum; 7a,7b— Khazar 
age of Volga valley, ChernyeYar, un-numbered, PIN [Paleontological Museum] ("Late" 
Elephas trogontherii and "Early" mammoth); 



421 



(423) 




да a 

FIGURE 175 (continued) 

Elephas (Parelephas)trogontherii: 8a, 8b- third terrace of Dniester at Tiraspol, 
un- numbered, Moscow Geological Prospecting Institute. Elephas (Archidiskodon) 
meridionalis: 9a - Araks valley , Leninakan, Kazachii Post, un-numbered, Geological 
Institute AN Arm. S.S. R. ; 10a, 10b - conglomerates of Taman Peninsula, No. 1249/222, 
No. 1249/42, PIN (Phanagorian ('. ? ) elephant); 



422 



(424) 




FIGURE 175 (continued) 

lib - conglonnerates of Taman Peninsula, No. 1249/232, PIN (Phanagorian С ? ) elephant); 
12a, 12b - Obitochnoe, Ukraine, No. 24239. ZIN; 13a - Stavropol, No. 25284, ZIN 



423 



(425; 

















































со 


оо 


LC 












о 














(М 




гН 


CJ 








О? 




со 






2 






05 
(M 


oq 

ОС 


со 

00 


о 


с 


^0 1 


со 

CJ 


1 


СО 

оо 


1 


со 


1 










































о 


О 


CD 


с 








^_^ 




со 


со 












'^ 


"7 


^ 


со 








о 


ст- 


со' 


со 




(M 


00 


Г- 


О 


ю 


ю 


1—i 


с: 


"* 1 СО 


1 1 


С- 


1 




со 


05 










(M 


00 


с^ 


rt' 


^ 


1 




'"' 


^ 


ос 


,-н" 














со 


О! 


Ч 


со 






^. 


ОС 


ст> 


■^ 












а> 


оо 












оо 


ОС 




со 








































^^ 












,-^ 




о? 
















о 












гН 
























CJ 












I— 1 








CO 


1 


00 


от 


1 


о 


t-H 


со r~t 


ю| 


с- 


1 


1С 


о 


00 










""^ 




00 


^ 


о 




^1 


'~' 


оо 


ст- 


^ 


























со 




















оо 




^ 








oj 




ю 


























'"* 




















































с? 


с 




ю 






















1 


^ 


^ 


'-<" 




(M 


^ 


CO 


(M 


(nI 1М 


о о 


со 1 со 


со 1 


^ 




о 


^' 


^ 










•"^ 


т—\ [ 1-Ч 


I-H I- 




*"* 1 


•"• 


^ 


1 


^ 






















ю 


с 


•"• 


•— ' 






















^" 






со 






















'"' 






'"' 


s: 




















































о 
E 


















о 


1 ^ 




оз 










1 


1 


1 r-i 


1 




1 ^ 


1 


о 


E 




















































<; 










1 


1 


1 


1 




1 
















1-< 










1н 
















О 












<и 














(U 




> 
о 








% 




> 

о 














E 




<L> 




;-. 




Ё 




S 


д 








-o 




С 




л: 




П) 


-о 


с 




^й 


П1 








(U 




(U 




о 




S 


U 






о 


и 








3 




о 




о 
о. 




(U 


"О 

3 


о 




о 


и 












>-< 




"и 




Ё 




0) 




<и 


Е 








J= 




л 




Ё 




с 


х: 


XI 




Ё 


П1 

с 








s 




Ё 




« 




S 


Ё 




П1 








(U 




a 




0) 




о 


й) 


3 

с 






о 




































о 




E 




о 


S Е 


о 


Ё с. 




о 


<л 


i 




,_ 




3 


*-■ 


5 Ё 

-О о 

Е о 


t^ 


3 Г 




5 Ё 

Ё о 


<и 




Ё 




E 


О 
О 




а) 
Е 


Е ^ 
2 с 




с 
о 


Е 
с 




3 




пЗ 


a. 


3 '-< 


j= с: 


3 


та Q 


L. 


3 ■-< 


5 


о 




z 




2 




Z 


н 


Z 


2 




Z 


н 






■ — 




— 




-V 




■^ • — 






-V 




— ■ 








-D 


li) 












^ 












С 


E 

3 


я 










и 










о 
о 


С 
> 

Q 
С 


3 

E 


^ 00 "^ 
о 3 -о 

г" >, о 

•^ S ОС 

- •^ 1^ 

« -Я 


-Ж 
00 






л 

р 

U 

э 


■;;; о 
._ 1) 










11 


1-4 

(L> 


с 
о 

-С 


о С -^ 


с 






> 1. 












IH 


JU 


§ 5 ° 
> > (Л 


л: 






и 


i2 Cl, 










о 


О 

Z 


3 


< 








3 3 










J 














D 


















Э§В >fSuXlBA4 


М Jaddn 
























э 


jaoojsi: 


>Id 














1 



с ст] 



424 



(426) 



) 
































































о 










ОТ 


СО 


























LQ 




OJ 








о 


оз 


гН 


<м' 










1Л 






OJ 


r-l со 


1 




00 


оо 

ОО 

со 

оо' 


«3 


00 


col -* 


OJ 1 

C<l| 


С<1 


1 

оо 
аз 
оо 


00 

ю 

00 


ОО 

оз 


оо 
о 

CSl' 


1 |со 


1 1 LO 

' |(М 




1 


аз 

о 
оо 




1 


оо 

о 










о 


1Г-_ 


ОС 


























CJ 








о 




^ 


^ 










аз 




о 














оо 


^ CD 






ю 
ел 


ю 

ОС 

со 


ю 

CD 


CD 


1 <N 


о 


о 


1 


оо 


1 


1 
аз 


■-н (N 


о о 




1-Н с- 




CD 

гН 


сЬ 










05 




























оо 




























































^^ 




о 








I-I 






ОЗ 




(N (M 






о 
оо 


с; 

аз 
и; 
с 


гН 

1 

10 


Ч 
Ч 


1 ::l 


1 


tr- 


1 


аз 
со' 


1 


<м' 
с^_ 


'*'| -и 


оо| ос 




U0 

оо" 
оо 

аз 


о 




со 
ю 














































^-^ 




^ 












у^ 




Я- 














ю 








CN 
















(N 




-l' 
















-;" 


CO <-< 


1 


CD 


О 


о 


о 


'~ 


1 со 


1 


СО 


1 


О 


1 


о 


1 оо 


1 2 




1 


о 




1 


00 

о 


















































гН 




CN 
















с< 














(М 




















































'"' 














'^ 
















1 ^ 


1 


IC 


1 2 


1 S 


1 






1 


1 


1 


1 




1 




1 




. 




. 


1 оо 
























'Г 


1 


О 


||2 


1 


О 


1 






1 


1 


1 


1 




1 




1 




































и^ 
















и 












(U 














0) 












u 




> 

о 








1) 




> 

о 










Щ 




> 

О 












s 




Щ 


д 






Ё 




5 




д 






Е 

as 




S 




я 




~o 


С 




^ 


cd 


■о 


с 




л: 






-а 


С 




л: 




л 




3 


о 




о 

о 


5 


а) 
-а 

3 


1) 

о 




о 

о 

Оч 






-о 

3 


О 




о 
о 
а. 








4-» 


<u 




U 


Ё 




<и 




и 




Е 




О) 




щ 




Ё 




^ 


л 




Ё 


п) 


х: 


X 




Ё 




с 

О) 


X 


X 




Ё 




03 




s 


e 




nj 




а) 


Ё 




сЯ 




0) 


Ё 




п) 




<и 




s 


3 

с 




0) 


О 


<и 


3 

с 




С 




о 


<и 


3 

с 




С 




о 




о 


E 




"о 


^О 


Е 




о 






о 


Ё ,. 




"о 




S Ё 
5 Ё 
3 с 


f-l 

E 


3 
Ё 


с 

с 


5 Е 

XI с 
Ё е 


u t 


X 

Ё 


3 

Ё 

X 




с 


5 Ё 

X о 

Е о 


5 

X 
Ё 


е1 

— ' и 

5 = 




S £ 

X с 

Ё о 




Э 


nj 


с 


L, 3 "- 


X с 


3 


СЧ 


с 


к 3 "-1 


х: -^ 


3 


1 ^ 


U 


3 '-< 




X CJ 


Z 


2 




Z 


Н 


2 


2 




Z 


Н 


Z 




Z 




н 






















































































п) 
































>> 








cd 














(1) 










^ У 






с 


X 
о 














tx 










> ■£ 






г^ 


3 














. Сц 










о ■- 








о 












^ 


В ^ 








а 


■? "о 






х; 


D 










с 


^ ^ 








1 


« .Н 13 






о 
о 


X 

и 


^ i 


с 










.S о 
ы си 








с 
D 
1 


7^ ^ °- 

0) ^^ >-■ 

Ё « ^ 






с 
5 

X 


0) 

Ё 


2 ;£ 


1 








"С 


2 о. 








тэ 


■-Da, 








> ^ 


о 








у 








5 


^ -S D 






о 


и 


0! rt 


2 








i 










2 








Z 




















э2в 


>fSU/(l 


елч>1 jaddn 


эЗв 


JBZB 


ЧМ 


























эиэ 


эо 


1ST3I 


i 


























1 



425 



(427) 



о 


с? 


00 


(N 


о 


OJ 


ю 


СТ1 


г- 


'-' 







^^ 


с- 


оо 


■-I 


ю 


ю 


со 


CD 


t^ 


г-|' 




I 2 




1-1 


3 


^ 


к-< 


^ 


0) 


h 




Е 


^ 


о 


^ 


с 


6 


6 


X 


^ 


Е 


^ 


о 


с 


3 




с 


Э 


CJ 


j=; 


с^ 


Z 


д 




>i 




н 
























а] 














>^ 












й 


п) 


























> ■;= 


il 


ей 


12 






J5 


*-• ct 


►-Н 











о до 



"^ й- >> 



^ > -::: 



(U Я "< 



»-( ПЗ пЗ 

Й " о 
-2 о н 



1-1 


м 


с:! 


(U 




> 

о 








ьо 


d 


О 


J3 


О 


« 



эЗе jpzBqM aippiiAi 



э§в rnjFa JaMoq 



3U9D0JS13Id 



426 



(428) 



I 3 





^-^ 




о 
































00 











ю 


(M 


CD 


CO 


ci 


IT- 


to 

-* 


CO 


CD 


(N 



о 




Ю 


05 


Ю 


(M 


•* 
1^- 




■* 





CO 


00 


CD 


CO 


U3 


oo 


Ю 


(M 



(U g 



S Ё 

E о 







(U 




> 




о 




s 


^ • 


JU 


ca 


о 


:г • 






ex 


0) 


(U 


E • 


e 


cfl 




(D 


0) 


О 



5 E 

JO о 

E о 



■П о 



г^ > N 



'^ 5 о -^ -о a, о 

^ ^ 00 ^ oj о о 

2 ^ о -J £ (Л о 

_o 

эЗв п>]вд 1эмот 
gugoojsiaid 






эЗв uojgqsdv Jsddn 



эиээоЦс] 



427 



(429) 

































ю 


LO Ю 


Ю 






ю 


Я S? 


;2г 




ю 


LO 00 


■*' 






ю 


^7 со" 


'^" 


s" 


п м" 2^ 1 2 ^' 


Ю CO 
Tf' CO 


о 
со 


OJ ■* 


«3 1 ю 

гН 1 ^Н 


^' 




ю 
со 




оо 










оо 




■^ ю 




-*' 


о .-1 
Ю CO 


со 






М"' 


со • 
со 


оо 
со' 






















05 








ю 


со 


















• 








to ^ 


со 








^ 


со' 


Y' 


1 1 "^ 


Ю 00 




1 |со 


1|^ 


1 


CS1 


(М 
















- 








Ю lQ 


OJ 








ч 


о 




со" 


со 








^' 


со' 




















со 






























CD 


со' 










2 


1 ^ 1 00 1 


ю 1 


^. 


1 |'-< 


1 1 00 


1 1 ю 1 1 '^. 1 






см 


1 






|со 






uo 
















Ю 


СП 














(N 
















1 




1о 


2 


1 1 1 1 




1 |оз 


ik Чоо- 




1 1 1 1 


1 


1 


1 


1 


2 






1 


1 


1 


^ -* 












2 


1 1 1 1 


1 


1 


1 


1 




























<L> 












0) 








,_^ 


> 








f-H 




> 








U 


О 








(J 




о 








: Е 


2 


^^ 






Е 




«2 








л 


HJ 


•— * 






л 




<и 


д 






-о с 


.a: 


cd 




-а 


с 




Jti 


пЗ 






<и <и 


о 


s 




и 


U 




о 






(U 


1 о 


о 

Си 


"u 




-о 

3 


"о 




о 
а. 




и 

"О 


"" 5 


u 


Ё 






0) 




щ 


Е 




с 


х: jo 


Ё 


с 




л: 


.о 




Ё 


«1 






S Е 


OJ 


<u 




S 


Ё 




п) 










с 


о 




W 


3 

с 




с 






















о Е <„ о • й 'Р 

5 g 5 BE ё 1 
Е ^ о Е о .Ус 


о 


Ё „ 


о 


(Л '"^ 




,_, 


3 t; 


S Е 


S Ё 




О) 


с "" 


5 е 




Ё 


2 о 


Е о 


.у с 




3 i5 D, 3 "-1 J2 W 


3 


« С1. 


3 --I 


JH d 




2 2 Z н 


2 


2 


2 


Н 








^ 






« 


'5^ 

1Л о . N 

U -^ .-- « О 

> 5 .. К ^ О 


'о 
> 


- > < 

° о - 


- О 

Л| О 
сЧ •- 

■^2 
о ^ 






2 > « •" л 


S 


(Я М п1 

§ < > 

с Е S 
Т S ^ 
2 -5 S 
S ё S 

а, с J 




.J 


Й) -г: -й О S -5 ^- 0, 


DO 

-о 
с 

■о 

с 


2 . 
- о 
^ а 
а. <ч 

J= я) 


^ 5 
о 

с 




oo 


со 








тз 












5 s 












эЗв \^Зщоц-\/ 


jaddn 








о t5 

(X 


auaooijd 










Щ 




























1 



428 



The above-mentioned morphological changes in the teeth of the supposed 
evolutionary sequence can be traced on the photographs of their longitudinal 
sections (Figures 173, 174) and by casts and diagrams (Figure 175). 
Table 91, which gives the morphometric indices of the teeth from some 
"key" burials in the Caucasus and the Russian Plain, is of practical 
significance in stratigraphic correlations. -^ 
430 The changes in tooth structure may be summarized as follows. From 
the time elephants appeared in the region of the Black and Caspian seas to 
the time of their extinction in the Lower Holocene, the maximum number 
of enamel loops of My increased from 16/15 to 25/19; the number of loops 
over 10 cm of the worn or lateral surface increased by 4.1/4.2. Thickness 
of the enamel wall decreased by 2.05/2.30 mm. 

The apparently saltatory changes in the tooth structure of Quaternary 
elephants, described so briefly in the preceding section, may be better 
explained from the geological (taphonomic) rather than the biological 
standpoint. The apparently isolated stages of morphogenesis of elephants 
(and other animals) may correspond to the changes in the structure of 
river networks and distribution, and in cycles of sedimentation which 
produced the conditions necessary for the burial of bones. Evolutionary 
inertia, which almost always operates in the evolutionary process, results 
in the lag of adaptive changes behind rapidly changing environmental factors. 
The accumulation of sediments depends, of course, on environmental 
factors; however, we tend to regard the evolutionary changes described 
as a continuous (not necessarily saltatory) accumulation of minor features, 
terminated by evolutionary extinction. 

Unfortunately, there are no data so far which would indicate that the 
development of mammoth features in the elephant stock occurred in the 
Caucasus at the same time as on the Russian Plain. Future comparisons 
of the results of geological and paleontological studies will undoubtedly 
reveal the truth. 



Order PERISSODACTYLA 

Very little work has been done on the phylogeny of perissodactyls, and 
as yet there is no agreement among paleontologists about whether the 
rhinoceroses of the Upper Pliocene and Anthropogene (Rhinoceros 
etruscus, R. mercki, R. t i с h о r h i nu s) represent a single 
phylor^enetic series. The pronounced morphological differences between 
these stratigraphically separated species are less sharp in the transitional 
stratigraphic stages. The skulls of the still poorly known 

R. binagadensis from the Middle Pleistocene of the Apsheron Peninsula 
(Binagady) differs from the Upper Pleistocene woolly rhinoceros in its 
deeply concave intertemporal region, the smaller second horn and greater 
inflation and callosity of the base of the first horn, the shorter postorbital 
part of the skull, thin mastoid processes and small (spherical) occipital 
condyles. The teeth of this species were considerably bigger (broader and 
longer) than those of the woolly rhinoceros, and the aboral edge of the last 

* In the construction of the table it was taken into account that the direct affinity of the Upper Pliocene 
forms to the Quaternary forms has not been proved in all cases. 



429 



molar protruded far beyond the anterior edge of the orbit (Figure 176, 2). 
The preorbital foramina are very small. The relatively narrow occipital 
region of the skull of the Binagady species indicates that its head was held 
in a more horizontal position than that of the woolly species. The Binagady 
rhinoceros probably thrived more on tree branches than on grasses, as did 
the woolly rhinoceros. The more slender metapodials, forearm bones and 
tibia of the Binagady rhinoceros indicate that it was more lightly built and 
was more active than the woolly rhinoceros. The abundance of this species 
in the clearly xerophytic landscape of the Apsheron, very different from the 
habitats of the woolly rhinoceros, attests to the distinct specific and even 
generic identity of the Binagady form (see Dzhafarov, 1955). All the data 
^•^^ presented above do not preclude the possibility that the woolly rhinoceros 
migrated from the north to eastern Transcaucasia and farther south into 
southwest Asia in Upper Pleistocene times. This seems very likely, as it 
is known that other large mammals like the primitive bull migrated in the 
Upper Pleistocene from the north to the Apsheron Peninsula. Since the 
remains of the woolly and early rhinoceroses in the Caucasus are 
represented by fragmentary material, at present we can only give a map 
of their occurrence (see Map 77), without going into a detailed analysis 
of their origin and stratigraphic variation of morphological and phylogenetic 
features. 




FIGURE 176. Rhinoceros skulls; 1-Rhinoceros t ic horh i nus , Upper Pleistocene , 
Russian Plain, Vladimir, No. 10699, ZIN; 2 - R h . b in aga d e ns is , Middle Pleistocene , 
eastern Transcaucasia , Binagady, No. 24402, ZIN 



430 



The evolutionary trends of Elasmotherium (E. caucasicum — 
E. fischeri), known only from its Upper Pliocene to Middle Pleistocene 
representatives, included a considerable decrease in the size of teeth and 
development of plicated enamel plates (Borisyak, 1914). 

The problem of the direct and indirect continuity of evolution of 
morphological characters in horses from the Anthropogene of the Caucasus 
and Russian Plain is as difficult as the problem of the evolution of 
rhinoceroses. The great ability to migrate and the almost universal 
adaptability to various types of habitat make it very difficult to solve the 
432 problem of the origin of the east European and Caucasian horses in the 
Quaternary satisfactorily. For example, it is still not clear whether 
Przewalski's horse is an evolutionary successor of the local Upper 
Pleistocene horse (broad-footed horse from the Upper Paleolithic of the 
Don area), or whether it migrated from the south in relatively recent times 
(i.e., Holocene). However, it is clear that local races of horses evolved 
in the Caucasian Isthmus and the Russian Plain, replacing one another at 
different geological periods. The general trend in horse evolution was 
reflected in the development of hypsodonty and monodactylism, but the 
details of the process varied from region to region. 

Considering the data presented above, it is only possible at this stage 
to speak of general stratigraphic and geographic changes in the post-Tertiary 
horses of Russia, without relation to the evolution of any particular 
species. As far as the Caucasian and east European material shows, the 
evolution of horses proceeded as follows. The Upper Pliocene horses of 
the Equus stenonis type from the Azov area were large, relatively 
heavy, with large teeth, characterized by a relatively thin and much folded 
enamel plate. The metapodia were neither highly specialized for fast 
running in semideserts and steppes nor broad-footed and thus adapted 
to the meadows and peat bogs of the taiga and tundra. 

The horses of the Upper Pliocene-Lower Pleistocene of the 
E. siissenbornensis type (from the conglomerates of the Taman 
Peninsula) had completely equuid teeth; their metapodia were lightly built, 
indicating that this form was adapted for fast running on savannah and 
steppes, which were regularly burnt during the summer. These horses 
were bigger than the Quaternary horses (Vereshchagin, 1957a), but a 
smaller horse occurred in the Middle Pleistocene of eastern Transcaucasia, 
characterized by more antero-posteriorly flattened metapodial diaphyses. 
Protuberances of the lower epiphysis ridge of the canon bone and 
metacarpus are indexes that the osteochondrous type of pastern fixation 
is somewhat less marked in the Binagady horse than in the Upper Pliocene 
forms. The laterial ligaments, however, are stronger. The hoof phalanges 
of this horse are relatively small, with a steeply rising angle anterior to 
the wall, an adaptation to the xerophytic habitats and the rough ground of 
this region. The fossil representatives of Upper Pleistocene and Holocene 
horses of the Caucasus are still poorly known; however, horses of the 
southern type, of "dry" contsitution (see Chapter III), undoubtedly inhabited 
Transcaucasia. The Upper Pleistocene horse of the Russian Plain was 
characterized by very massive legs and a brood foot, which can be seen 
particularly well in the representatives from the Upper Paleolithic beds 
at Kostenki on the Don (Figure 177). 



431 



(4зз; 




FIGURE 177. Stratigraphic and geographic variation in the shape and supporting surface area 
of the hoof phalanx of Eurasian horses, a — fore legs; b — hind legs. 2 natural size 

la.lb — Equus caballus gmelini, Recent "tarpan", Russian Plain, No. 521, ZIN; 

2a,2b — E. przewalskii .Recent Przewalski's horse, Dzungaria, No. 17591, ZIN; 

3a,3b — E. caballus subsp, (foss.). Lower Holocene, Novosibirskie Islands, No. 4419, ZIN; 

4a,4b — E. caballus latipes. Upper Pleistocene (Upper Paleolithic) Voronezh, Kostenki XV, 

No. 25183, ZIN; 5a,5b— E. caballus subsp., Middle Pleistocene, eastern Transcaucasia, 

Binagady 



432 



These horses probably grazed on soft meadows, tundra and taiga swamps 
on the floodplains of rivers. A similar structure of the hoof phalanx, though 
not as wide, is characteristic of the horses of the extreme northeast of 
Siberia, which probably lived on somewhat harder terrain under conditions 
of thinner snow cover. * 



434 TABLE 92. Stratigraphic and geographic variation of the supporting surface (mm^) of hoof phalanges 
of horses from the Anthropogene of Asia 







л 
































2 OO 




T3 <л 






„ 




1 




с 


3 3 




(U 






!)"§=> 






3 




nj 




u 


-3 ^ 




с 






с Д -' 














о 


CTJ ""• 




(U 


CO r— 1 




<U /" '-' 










S 2 




о 


о rt 




о 


3 "> 




о " П1 






й С 




1) iTi 


Dimensions 




с -^ 


w 


о 


t^ л 


en 


2 .ii -^ 


со 




-° < 5 




N f^ 




Q) 


a л 


с 




- л 


с 


." -^ П1 


С 




-S^- 








CU 


P ^" 


a 


^ 


§ " ^ 


(0 
6 


5 й о 


0) 




.2 


^^- 55 




XJ 


»li 


ol 


Q и I" 


^ § и 


щ 


ел '^ i-H 


ifl 


CO 2 ^Л 




о 


1-4 


M 3 Oh 


О 


*м ел -3 

4) о ;! о, 


'о 

D 


с 




DO 

С 


3 ^ ^ 




13 


tj nj 3 


a. 


Оч 


Q. 3 ■;; 


&, 


о. > 3 со 


а, 


м 


О- Ё О 


3 


3 .^ • 




•^ 


Й .5 a- 




D. 


& O" ce 




а, о о- ■§ 




Л 


N 


°* л о 




2 


(U 03 Ы 


■* 


P 


3 W -4 


■* 


р 2 ы S 


со 


D 


W DO Z 


Q 


w .л 2 






28.8 
27-32 






43 




40 






34 




37.5 


Fore leg 


39-50 


27-48 




Hind leg 




25.5 






40.9 
37-46 




37 






31.5 




33.5 


33.5-40.5 




Index of difference 


























(2:1 X 100) 




88.5 






95,0 




92.5 






92.6 




89.3 



Note. Means in the numerator, observed ranges in the denominator. 



The size of the area of the fore and hind leg hoof phalanges of different 
horses and their interrelationships are given in Table 92. The data in 
Table 92 show that in the southern and steppe varieties of horses the 
anterior hoofs were relatively larger. 

During the period of maximum cooling and development of mesophytic 
landscapes, the broad-footed horse migrated far southward, probably to 
the foothills of the Caucasus. There are reasons to believe that the 
xerothermal stage and the extensive development of steppes on the Russian 
Plain resulted in further divergence of these Upper Pleistocene horses in 
the development of morphological characters related to forest and steppe 
tarpans; confirmation is difficult, however, due to the absence of good 
skeletal material of these so recently extinct forms (see Gromova, 1949; 
Geptner, 1955). 



Order ARTIODACTYLA 

The large amount of work done on this order allows easy tracing of the 
miain evolutionary trends of some of the Caucasian forms. 



* A. A. Sludskii (pers. comm. ) related the occurrence of wide hoofs on the fore legs of the northern and 
Upper Pleistocene hoofed mammals to the need to dig through the snow in search of food. 



433 



Pigs of the group Su s scrofa — S. vittatus appear in Eastern 
Europe and the Caucasus for the first time in the Upper Pliocene, and 
their main morphological features by then have already been eatablished. 

Amon (1938) noted the extreme variation in skull morphology of the 
Pliocene -Pleistocene pigs, which already inhabited vast areas of Eurasia 
by the Lower Pleistocene. Therefore there are reasons for believing that 
any stratigraphically younger pigs found in Russia are members of a 
continuous evolutionary lineage. 

The Taman boar (S. tamanensis) from the Upper Pliocene beds of 
the Taman Peninsula differs from the younger Caucasian boars in its 
435 larger size and, in mature males, a much wider lower jaw in the region of 
the diastema (Vereshchagin, 1957a). The boar from the Middle Pleistocene 
beds of the Apsheron Peninsula (S. a p s с h e r о n i с u s) does not differ in 
size from the Recent species, but the Apsheron species is characterized 
by auditory bullae elongated downward and by the lachrymal bones of the 
type intermediate between S. scrofa and S. vittatus (Burchak- 
Abramovich and Dzhafarov, 1948). Widths of the lower jaw of Quaternary 
boars measured behind the tusks are given in Table 93. 



TABLE 93. Changes in size of boars by the dimensions of the lower jaw (in mm) 



Locality and geologic age 


Species 


Width of lower 
jaw behind tusks 


Number of 
specimens studied 


Caucasus, Recent 


Sus scrofa attila ... 


46.0 
45-55 


5 


Middle Don, 10-13th centuries A. D. 


S . s с ro fa attila 


59.0 
55-64 


3 


Volga area, Tunguz Peninsula, 








Middle Pleistocene 


S. cf. scrofa 


65.0 


1 


Transcaucasia, Apsheron 




52.0 


2 


Peninsula, Middle Pleistocene 


S.apscheronicus .... 


50-54 


Ciscaucasia, Taman Peninsula, 








Upper Pliocene 


S. tamanensis 


73.0 


1 



Note. Means in the numerator, observed ranges in the denominator. 



A number of lower jaws and skulls of boars of the Anthropogene are 
shown in Figure 178. The relative size of the bones of the axial and visceral 
cranium has changed considerably since the Middle Pleistocene; however, 
there have only been minor changes in tooth size since that time. The sizes 
of Мз are identical in the Binagady and Recent Caucasian boar. However, 
the heel portion of the tooth of the Recent species has become nnore complex 
due to development of 1-2 columnar papillae on the heel (see Figure 178, 1,4). 

The abrupt saltatory decrease in the size of teeth, particularly noticeable 
in Мз, occurred only in the Neolithic in domesticated populations. 



434 



The remains which have been reliably identified as the red deer group 
(Cervus elaphus s. lato) occur in the U. S. S. R. for the first time in 
the beds of the Lowermost Anthropogene. * These deer evidently evolved 
from some Pliocene Mediterranean forms. Pavlova (1926) has described 
a skull and very simple unbranched antlers of a deer (Cervus 
tschelekensis) from the Akchagyl (Upper Pliocene) beds of the 
Transcaspian region. Isolated fragments of antlers (with the first outgrowth 
somewhat above the crown) of a small variety of Pliocene deer occur in 
436 the Lower Pleistocene conglomerates of the Taman Peninsula. Their 

identification with the red deer group is not certain (Vereshchagin, 1957a), 
though it seems possible that they are ancestral forms. The Lower 
Anthropogene deer, known from the Tiraspol gravels (on the Dniester), 
are characterized by antlers with two closely spaced low supraorbital 
outgrowths and by the primitive, crownless type of the terminal branching. 




FIGURE 178. Skulls and lower jaws of boars 

la, lb — S us s его fa at tila , Recent, Greater Caucasus, No. 10485, ZIN; 2 — S.scrofa attila, 
10- 13th centuries A .D. , Lower Don. Sarkel; 3a, 3b — S . a ps ch er on i cu s , Middle Pleistocene, 
eastern Transcaucasia, No. 26031, ZIN; 4— S. tamanensis. Upper Pliocene, Caucasus, Tannan 
Peninsula, No. 26000, ZIN 

This type of "tightly" branching antler was remarkably stable, persisting 
through the Anthropogene in Eastern Europe and the Caucasus. However, 
the Middle and Upper Pleistocene deer of the Volga valley and areas farther 
east are characterized by a "looser" position of the first outgrowths on the 

* Indications of earlier occurrences, as for example, in the Middle Pliocene of Transcaucasia (Bogachev, 
1938c), are not cenain. They probably refer to the deer of the genus Eucladocerus. 



435 



antler stem similar to that of the Recent Siberian and American wapiti. 
Development of the fairly large antlers in the European noble deer began 
probably in the Middle Pleistocene. A weakly-developed type of vestigial 
crown is noted in some antlers washed out of the Khazar alluvium of the 
Volga (Figure 179) and Don valleys, and the crown is well developed in 
some antlers from the bitumens of the Apsheron Peninsula. 



(437; 




5 10 IS 20 25 cm 

'■■■■■ 

FIGURE 179. Antlers of red deer (C e г vus elaphus subsp. ) from the Middle 
Pleistocene alluvium of the proto-Volga valley. Kuibyshev museum 

The extensive descriptions of deer from the Binagady materials 
(Alekperova, 1952) indicate that the Middle Pleistocene deer of eastern 
Transcaucasia are not appreciably different in size from the Recent deer 
437 of the Greater Caucasus. The Middle Pleistocene species is characterized 
by lighter antlers. According to Akeperova's drawings, antlers with poorly- 
developed and widely-spaced supraorbital outgrowths also occur in the 
bitumen. This material is not sufficient for the study of the role of the 
migrations and local speciation of the noble deer of Eastern Europe and the 
Caucasus in the Anthropogene under the conditions of changing landscape 
and ecology. Nevertheless it is clear that during the Quaternary in Eastern 
Europe and the Caucasus there was a "struggle" between the elaphus - 
type (Cervus elaphus — with developed crown) and the wapiti-type 
(C. canadensis — without crown and with widely-spaced first outgrowths). 
For the Caucasian population the development of this process can be traced 
by using Dinnik's (1914a) descriptions and very complete collections 
as those of the Digorized cave in North Ossetia and the museum of the 
Borzhomi nature reserve. The elaphus -type of antler is usually 
suppressed in the Recent Caucasian deer, but in the Digorized collection, 
which in 1948 consisted of 62 pairs and 121 single horns, 27% were 



436 



e 1 a ph u s-type antlers (Figure 134) and 73 % wapiti -type.* The reliability of 
these estimates is somewhat questionable due to the fact that many antlers in 
this collection have an intermediate type of terminal branching, i.e. , 
vestiges of a small crown. 

Smirnov (1923-1924) pointed out that the Caucasian deer has decreased 
in size since the Upper Paleolithic, his conclusions being based on the 
occurrence of teeth in the Gvardzhilas cave. Indeed the teeth found there 
by us and D.M. Tushabramishvili are of striking size, almost like those 
of elks. 
438 Since the collections contain mostly Quaternary deer material, it is 

only possible to speak in greater detail of stratigraphic variation in antler 
size. The antlers, and probably the animals themselves, gradually 
increased in size from the Lxjwer to the Upper Pleistocene, the largest 
example occurring in the Upper Paleolithic of the Russian Plain (Kostenki, 
near Voronezh). Later, in the Holocene, they somewhat decreased in size 
again, though not as sharply as bison, for example; in the Middle Ages the 
deer of the nniddle Don valley (Fortress Sarkel, near Tsimlyanskaya) were 
still very large (Figure 180, 181). It is of interest that size change 
in deer somewhat lags behind such changes in bison (see below). 

The stock of particularly large individuals which colonized the Russian 
Plain at the end of the Upper Pleistocene was probably derived from the 
Caucasian population of deer which survived the Quaternary cool phases 
under most favorable conditions. The migration of the Caucasian deer to 
the north probably followed the Don and Volga valleys. In any event, 
Flerov's (1952) statement that they attained their maximum size in the 
Middle Pleistocene (Mindel-Riss) and then gradually began to decrease in 
size is undoubtedly wrong. 

The Recent Caucasian deer are characterized by body dimensions and 
size of antlers intermediate between the West European and Asian- 
American forms. This has been correctly noted in the first description 
by Ogilby. 

Geographic variation of the Recent deer C. elaphus maral in the 
Caucasus has not been studied in detail; variation is to a large extent 
masked by individual and age variability. 
440 According to Dinnik (1914a), the deer of the northwestern Caucasus 
(Kuban area) are the largest, their size approaching that of Manchurian 
wapiti. The size of the Caucasian deer decreases to the east and south, 
and those inhabiting the foothills of Dagestan and the tugaic lower reaches 
of the Terek and Sulak are particularly small. Both the fossil material 
and the Recent populations are characterized by light horns, with only a 
few (5-6) outgrowths, and almost always devoid of crown. The deer which 
inhabited the Armenian Highlands and Talysh were also of relatively small 
size. All this is accounted for by the optimum conditions of the northwest — 
abundance of food, vast areas of forests and mountains, etc. — as compared 
with the hot and dry east and south with their ancient civilizations. 

Thus it can be seen that the picture of the development of the 
morphological type and range of distribution of the red deer in Eastern 
Europe and the Caucasus in the Cenozoic is extremely complex. This also 
shows that some of the excellent paleozoogeographic analyses of Geptner and 
Tsalkin (1947) can be confirmed or disproved only by means of additional 
collecting and detailed study of paleontological materials. 

' Dinnik's statement on the 500 pairs of horns at Digorized is somewhat exaggerated. 

437 



(438) 




о .2 

■ёа 

о о 
Д о 



с 5 N 

Щи. 






Е 



Ё,^ 



> а 



^" Е Б 



<и --• (1) ГТ! 



л 


> 


о 


о 


0) 


а 
о 
Q 


2 


о 


с 


<и 


о 


cd 






s? 






S 


о 


"о 


Q 


U 


2 


12 

с 

Е 


< 


-а 
■а 


-о 


« 


2 




00 

ш 


3 


1 


Е 




с 


•* 


3 


о 


о 


> 


1 
с 

3 


оо 


(й 


X 


ы 


1 


12 


о 


а; 


о 


с 


р, 


D 


'"' 


Щ 


ш 


О 


1 


о 


*-* 



438 



(439; 




30 33 40 45 50 5S 60 OS 70 75 mm 

FIGURE 181. Change in antler size in red deer of the Cervus elaphus group. One- half of the sum of 
the diameters of the antler bases under crown. Black dots indicate means. Ordinate - frequency 

1- 15-19th centuries A.D., North Ossetia, Digorized cave; 2- 10- 13th centuries A.D. , Don valley, 
Sarkel; 3 - 1st millennium B.C. , Caucasus, burial, Makhachkala, Samtavro and Lake Sevan shores; 
4- Upper Pleistocene (Upper Paleolithic). Don valley, Kostenki XIV; 5 - Middle Pleistocene, bitumens of 
Apsheron Peninsula, Binagady; 6 — Middle Pleistocene, alluvium of proto- Volga; 7 — Lower Pleistocene, 
alluvium of proto-Dniester, Tiraspol; 8 — Upper Pliocene, Caucasus, Taman Peninsula 



439 



The phylogenetic lineage of the Recent elks of the genus Alces is 
so far known only from the Middle Pleistocene. Nevertheless, there is no 
doubt that Alces alces evolved at least in the Upper Pliocene, since 
by the Middle Pleistocene the animals were already indistinguishable from 
the Recent forms. The elk-like cervids of the side lineages which became 
extinct in the Lower Pleistocene, Tamanalces and Alces latifrons, 
had, by the Upper Pliocene, developed some features of skull and antlers 
similar to those of elks. These features developed, evidently, long before 
the Anthropogene. 

The adaptations of the Recent northern elk to life in the taiga swamps 
and deep snow (long legs, large supporting surface of hoofs, brachydonty 
and feeding on twigs) could have originated long before the elks invaded 
the taiga and moss swamps. - Until recently, elks inhabited (in 1951) the 
floodplain marshes of the middle Volga area. Thus, Flerov's (1950, 1952) 
statements on the recent origin of these adaptations and on the feeding of 
elks mainly on leaves and bark do not correspond with the facts. 
It is known that for 7-8 months of the year elks feed mainly on hard 
(frozen) twigs. 

The skulls and antlers of Middle Pleistocene elks from the Khazar beds 
of the Middle Volga region were already practically indistinguishable from 
the Recent forms. The Caucasian elk, A. alces caucasicus, known 
in the Caucasus since the Upper Pleistocene, was a relatively small animal 
in the 18th century with clearly pronounced features of the European form. 
The skull is relatively broad, particularly in the preorbital and frontal 
regions; the nasal section is narrower than in the northern elks. The 
molars differ in their branchydonty with complicated wedge-shaped folds 
on the labial surfaces of Pm^ and Pm^. In general, the morphological 
features of the Caucasian elk are sufficiently well pronounced, which enables 
one to date its appearance and beginning of isolation in the Caucasus as 
at least the Middle Pleistocene. 

The saiga (Saiga tatarica) is an example of considerable 
morphological stability since the Middle Pleistocene. However, the Russian 
and West European collections consist mostly of teeth and frontal- 
442 occipital fragments with horn cores . The oldest and probably most primitive 
type of the frontal-horn fragment was collected in the Khazar beds 
of the central Volga area. The skull of the Middle Pleistocene East 
European saiga is relatively small, with absolutely straight, fairly 
short and only slightly diverging horns (Figure 182, 4a, 4b), quite 
similar to the skull of the saiga from the Binagady asphalt in 
eastern Transcaucasia. While no different from the Khazar skull type in 
the divergence of the horns, the horn stems of the Binagady skulls are 
longer, with a noticeable posterior curvature in the sagittal plane 
(Figure 182, 3a, 3b). -* 

Upper Pleistocene skulls from the Khvalynsk beds of the Ural valley 
(our collections, 1950) are characterized by greater divergence of the 
horns, curvature of the horn stems in the frontal plane, massive structure 
in some specimens and even greater length in other specimens 
(Figure 182, 2a, 2b). 

* It should be mentioned that all these features developed also in the absence of snow in Tertiary giraffes and 

in some antelopes of the Mediterranean and Africa. 
•• Unfortunately, Alekperova's (1955) detailed description of the Binagady saiga does not help in drawing 
conclusions. 



440 



(441) 









FIGURE 182. Stratigraphic and geographic variation in structure of horn cores and 
frontal- occipital fragments of saiga skulls. 1/5 natural size 

Frontal and side view: la, lb — Saiga borealis Tscher. , Lower Holocene, Yakutia, 

Vilyui valley, No. 13636, ZIN; 2a,2b-S. tatarica. Upper Pleistocene (Khvalynsk 

age), alluvium of proto-Ural, No. 24201 (10), ZIN; 3a , 3b — S . tatarica b ina gad e ns is , 

Middle Pleistocene, bitumens of the Apsheron Peninsula, Blnagady, No. 22385, ZIN; 

4a, 4b — Saiga sp., Middle Pleistocene (khazar age), alluvium of proto- Volga, No. 1084, ZIN 



441 




FIGURE 183. Outline of hom srems and frontal-occipital section 
of skull of the Recent saiga 

1— Saiga tatarica, Volga-Ural steppes, No. 7064, ZIN; 
2 — S.tatarica rnongolica, northwestern Mongolia , 
No. 8465, ZIN 



This type of frontal-occipital has persisted from the Upper Pleistocene 
to Recent times in vast areas of Eastern Europe, central Asia and west 
Siberia. In Yakutia, however, on the Vilyui and Lena rivers, 
morphologically distinct populations of the northern saiga (S. borealis 
Tscher.) are known from the Upper Pleistocene and Lower Holocene, 
characterized by the elongation of the region posterior to the horns, and 
straight and long laterally compressed horns with weak divergence 
(Figure 182, la, lb). 

The frontal-occipital region of the skulls in Recent saigas shows wide 
variability; however, their skulls are practically indistinguishable from 
the fragments of skulls from the central Urals (Figure 183, l). 

The geographic variation in Recent populations is probably related to 
their occ