<<

Gennady Bary s h n i kov Zoological Institute, R A S , S t . Petersburg

C h ro n o l ogical and ge og r aphical variability of Crocuta spelaea ( C a r n i vo r a , H yaenidae) fro m the Pleistocene of Russia

B a r y s h n i k o v, G., 1999 - Chronological and geographical variability of C rocuta spelaea ( C a r n i v o r a , Hyaenidae) from the Pleistocene of Russia - in: Haynes, G., Klimowicz, J. & Reumer, J.W. F. (eds.) – MA M M O T H S A N D T H E MA M M O T H FA U N A: ST U D I E S O F A N EX T I N C T EC O S Y S T E M – DEINSEA 6: 155-174 [ISSN 0923-9308]. Published 17 May 1999.

Geographic variation in C rocuta spelaea dentition, beginning from the Middle Pleistocene, can be seen as specialization in western and eastern Eurasia. The sizes of C. spelaea increase from the south to the northwest and northeast. The of the Primorski Krai had the largest teeth.

C h ronologische en geografische variatie in Crocuta spelaea (, Hyaenidae) uit het Russische P l e i s t o c e e n – Geografische variatie in het gebit van de grottenhyena, vanaf het Midden Pleistoceen, wordt beschouwd als een specialisatie in westelijk en oostelijk Eurazië. De maten van de grottenhyena nemen toe van het zuiden naar het noordwesten en noordoosten. De hyena van Primorski Krai had de grootste gebitselementen.

Correspondence: Gennady Baryshnikov, Zoological Institute, Russian Academy of Sciences, 199034 St. Petersburg, Russia

Keywords: C rocuta spelaea, Russia, geographic variability, Pleistocene

I N T RO D U C T I O N The cave hyena C rocuta spelaea (GO L D F U S S , of bone assemblages of C. spelaea f r o m 1823) from Pleistocene Eurasia is usually European cave sites basically differs from regarded as an extinct northern subspecies of that in , permitting us to assume diff e- the recent C. cro c u t a (ER X L E B E N , 1777) rences in behavior between C. spelaea a n d (Kurtén 1956, Werdelin & Solounias 1991). C. cro c u t a (Brugal et al. 1997). The cave This assumption, based upon a formal simi- hyena was probably an active predator of larity of skeletal features of both forms, is in Pleistocene ungulates and may have lived in fact debatable. The view that C. spelaea is a l a rger groups with a complex hierarchical separate species is supported by the more s t r u c t u r e . carnivorous specialization of its dentition, as compared to C. cro c u t a (Baryshnikov 1995, An ancestor of C. spelaea was C. sivalensis Baryshnikov & Averianov 1995). One also (FA L C O N E R E T CA U T L E Y, 1868) from the may assume that the cave hyena had adapta- Siwalik fauna of India, which during the tions to life in a climate that had a long Cromerian interglacial spread into the tempe- frosty period (adaptations predicted, for rate latitudes of Eurasia (Kurtén 1956, 1957). example, to be in features such as the deve- I refer to C. spelaea all Pleistocene discoveries lopment of a dense fur coat). The character of the genus C ro c u t a from Europe, western

155 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Figure 1 Localities with Pleistocene remains of Crocuta spelaea in Russia and adjacent terri t o ri e s : a - Cromeri a n - M i n d e l , b - Riss- W ü rm , c - W ü rm . 1 - Kazanka Rive r, 2 - Binagady, 3 - Golabiec, 4 - Prolom 2, 5 - Ilskaya I, 6 - Nasorog Cave, 7 - Smelov s k aya 2 C ave, 8 - A m a n - K u t a n , 9 - Kra s nyi Ya r, 1 0 - Bukhtarm i n s k aya Cave, 1 1 - Ust-Kanskaya Cave, 1 2 - Denisova Cave, 1 3 - Geographical Society Cave.

Asia, middle Asia, Siberia and China; the on the localities of the last interglacial (Riss- genus evolved several subspecies replacing Würm) in Moldova (Vykhvatintsy), Ukraine each other in time and space (Kurtén 1956, (Iliinka, Kiik-Koba) and A z e r b a i j a n 1965, Bonifay 1971, Schutt 1971, Kurtén & (Binagady). We have a better knowledge of Poulianos 1977). the distribution of C. spelaea in the Early- Middle Würm (Fig. 1). In deposits of that In the former Soviet Union the earliest record period its abundant remains occur jointly of C. spelaea falls in the Cromerian/Mindel with a complex of Mousterian stone imple- transitional period. During that time interval ments. Cave hyena was found in Moldova, the remains of C ro c u t a are known from the Crimea, the Northern Caucasus, the Southern Ukraine (Opolie, Sinyakovo), Russian Plain Urals, middle Asia, Kazakhstan, and (Kazanka River), Northern Caucasus Southern Siberia eastwards up to the A n g a r a ( Treugolnaya Cave), and Kazakhstan River basin, while further in the east it appe- (Zhelesinskoe; Baryshnikov & Ve r e s h c h a g i n ars again in the southern Primorski Krai 1996). No significant discoveries are known where it penetrated from China (Baryshnikov for that area for the period from the Late & Vereshchagin 1996). In the Crimea and in Mindel up to the Riss, although remains of Middle Asia C. spelaea disappears by the C ro c u t a from Kyrgystan (Sel-Ungur Cave, beginning of the early Paleolithic, but in the Mindel or Mindel-Riss) and Moldova (Starye Urals and the Altai it could have survived D u r u i t o r y, Riss ?) could belong to the same until the main climatic minimum of the Late time interval. More information is available P l e i s t o c e n e .

156 BARYSHNIKOV: Pleistocene Crocuta from Russia

During the Pleistocene European C. spelaea Russian Academy of Sciences, Novosibirsk demonstrated notable differences in the sizes (I A E); the Institute of A r c h a e o l o g y, Ukrainian of skull and teeth, observed when the disco- Academy of Sciences, Kiev (I A U ); the veries from deposits of glacial and interg l a- Institute of Systematics and Evolution of cial epochs are compared (Kurtén & , Krakow, Poland (I S E ); l'Université Poulianos 1977, Klein & Scott 1989). Claude Bernard Lyon 1, Villeurbanne, France H o w e v e r, the tendencies of geographic varia- (U C B); l'Université Pantheon-Sorbonne, Paris, tion at different temporal levels of the France (U P S ); and the Laboratoire de Pleistocene are not sufficiently studied. In Geologie du Quaternaire, Marseille-Luminy, this paper the major tendencies of spatial and France (L G Q ) (Table 1). temporal variation of C. spelaea are discus- sed for the former Soviet Union and We s t e r n R u s s i a Europe. Most attention has been concentra- KA Z A N K A RI V E R. Complete skull with lower jaw from ted on molars, the sizes and proportions of the Kazanka River in Tatarstan (G M M 191); Middle which varied depending not only on the size Pleistocene (Cromerian-Mindel). of animals, but also on dietary specialization. IL S K AYA 1. Postcranial bones and lower cheek teeth (ZIN 19889). Mousterian open site at Il River (Kuban M AT E R I A L S Basin), Northern Caucasus; Early W ü r m . The materials studied include sixteen samp- NO S O R O G CAV E. Left mandible (ZIN 34477). A cave in les of fossil hyena remains from the collec- the South Urals; apparently dated to the Early W ü r m . tions of the Zoological Institute, Russian SM E L O V S K AYA 2 Cave. Approximately 45 C ro c u t a Academy of Sciences, St. Petersburg (Z I N ) ; remains (ZIN 34491).Upper Paleolithic site on Malyi the Geological and Mineralogical Museum of Kizyl River, South Urals. C ro c u t a remains come Kazan University, Kazan, Russia (G M M ); the from layer 5, lying beneath the cultural layer. Institute of Archaeology and Ethnography, Radiocarbon dates are: layer 3-4: 31,000 +/- 1500 y B P (GIN-8401); 41,000 +/- 1800 yBP ( G I N - 8 4 0 2 ) . KR A S N Y I YA R . Left mandible (ZIN 32752) from Ob River beach near village Krasnyi Yar in Tomsk Ta ble 1 The number of specimens studied. Region (Fig. 2). US T- K A N S K AYA CAV E. Postcranial elements and isolated teeth of hyena (ZIN 34404) within highly fragmen ted ungulate bones, in a Mousterian cave site on Charysh River in Altai; Karginian interglacial Mousterian deposits. DENISOVA CAVE. A few isolated bones and teeth (coll. IAE) from Mousterian (layers 12-20) and Upper Paleolithic (layers 9-11) site located on Anui River in Altai. Layer 21 has termoluminisciences dates 155,000 ± 31000 (RTL-546) and layer 14 as 69,000 ± 17000 (TL-9), and layer 11 radiocarbon date as >37,235 yBP ( S O A N - 2 5 0 4 ) . GEOGRAPHICAL SOCIETY CAVE. More than 170 bones and teeth, including fragmented skull (Fig. 3) and few lower jaws (ZIN 34478-34490) mainly from a depth of 80-140 cm below the surface (layer 4); Upper Paleo- lithic site located at Partizanskaya River (= Suchan River) in southern Primorski Krai. Bones of horse and mammoth from a depth of 60-80 cm were dated by radiocarbon as 32,570 +/-1510 yBP (IGAN-341).

157 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ukraine Fr a n c e PROLOM 2. 418 remains of hyena including fragments of LU N E L- V I E L . Numerous remains of cave hyena, including lower jaws, limb bones and isolated teeth (coll. IAU). whole skulls and mandibles (LGQ 334, 368, 736, Mousterian cave site at the left bank of Biyuk Karasu 2408, 2677, 2898, 5189, 12598); cave site is situated River near Belogorsk City, Crimea; Early Würm. nearby Montpellier (Herault) in the south of France; Azerbaijan M i n d e l - R i s s . BINAGADY. Numerous bone remains of hyena skulls with- JA U R E N S . Hyena remains studied by the author are not out lower jaws (ZIN 24407) (Fig. 4) and 10 mandibles abundant (UCB 300568, 300576, 300578, 300581, (ZIN 22382). Tar pit near Baku City; Riss-Würm 300582, 300586); Cave located in the central part of (Eem). France (Correze); Würm. Uzbekistan AR C Y-S U R- C U R E . Numerous remains include isolated AMAN-KUTAN. More than 55 fossil remains of hyena (ZIN teeth and bone fragments mainly from Mousterian 25217, 29058); Mousterian cave located on the western deposits from Grotte Renne (levels XI-XIV) and slopes of Zeravshan Ridge; Early Würm. Galerie Schoepfin (coll. UPS); several caves situated Kazakhstan in the central part of France (Yonne) yielded BUKHTARMINSKAYA CAVE. Small collection of fragments of Mousterian and Upper Paleolithic tools; Early and bones and teeth (ZIN 34403); Upper Paleolithic cave Middle Würm. Level XII in Grotte Renne has a site situated in Bukhtarma River mouth; Middle Würm. radiocarbon date of 37,500 +/- 1600 yBP. Some teeth

a

b

Figure 2 Left lower jaw of Crocuta spelaea from Kra s nyi Yar in We s t e rn Siberia (coll. ZIN 32752). Labial (a) and lingual (b) views.

158 BARYSHNIKOV: Pleistocene Crocuta from Russia

were unearthed from lower levels of Upper M E T H O D S Paleolithic deposits in Grotte Renne (IX-X). The maximum length (L) and maximum Po l a n d width (W) of upper and lower cheek teeth GO L A B I E C . A series of molars and bones (ISE MF/ were measured. In addition, the length of the 37/60); cave site in the southern part of Poland; metastyle (Lm) was measured on P4, and the Würm. length of the trigonid (Lt) on m1. Statistics include calculation of means (x) and stan- Collections with small samples from some dard deviation (SD). I also used Component sites were placed into two groups: ‘South Analysis and Cluster Analysis of samples on U r a l s ’ with the teeth from Smelovskaya 2 euclidean distances with hierarchy dendro- Cave and Nosorog Cave, and ‘Altai’ - from gram by the average linkage method in the Ust-Kanskaya Cave, Denisova Cave and statistical software SYSTAT. Bukhtarminskaya Cave.

a b

Figure 3 Cranium of Crocuta spelaea u s s u ri c a from Geographical Society Cave, Russian southern Far East (coll. ZIN 34478, h o l o t y p e ) . D o r sal (a) and palatal (b) views.

159 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

a b

Figure 4 Cranium of Crocuta spelaea from Binagady in eastern Transcaucasia (coll. ZIN 24407). D o rsa l (a) and palatal (b) views.

R E S U LT S during the Mindel-Riss interglacial (C. spela - C r a n i u m ea intermedia BO N I FAY, 1971). Next there European populations of C. spelaea r e v e a l was another insignificant increase (245-260 fluctuations of average values of condyle mm) in the Riss, another decline (236-262 basal length of the skull during the Pleisto- mm) in the Riss-Würm interglacial and, cene (Kurtén & Poulianos 1977). An increase e v e n t u a l l y, the increase of the length up to of condyle basal length from a large (267 its maximal values (240-288 mm) in the mm) in the Cromerian and Early Mindel (C . Würm glacial (C. spelaea spelaea GO L D F U S S , spelaea praespelaea SC H Ü T T , 1971) to a very 1823). Therefore, the size of the individual l a rge size (279.5 mm) in the Mindel (C. spel - animals was essentially different in the cold aea petralona KU RT É N , 1977) is followed by and warm climatic phases of the Pleistocene. an abrupt decline of its values (240-251 mm) Isolated data on condylar basal length in

160 BARYSHNIKOV: Pleistocene Crocuta from Russia

Ta ble 2 Geographical variability of length of P1- P4 (in mm) in Ta ble 3 Measurements (mm) of fi r st upper premolar P1 in Crocuta spelaea. ( N o t e : in Ta bles 2-6 and 8-12 C. spelaea Crocuta spelaea. p ra s p e l a e a should read C. spelaea pra e s p e l a e a . )

Petralona Cave (Schütt 1971, Kurtén & Poulianos 1977).

P 1 This relatively small tooth shows consi- derable variation in its proportions, which is typical of teeth not participating significantly from other regions of Eurasia, such in occlusion. It is relatively smaller in as from the Kazanka River in the Russian C. spelaea in comparison to the tooth in C . Plain, of Cromerian-Mindel age (268.5 mm), c ro c u t a (Baryshnikov 1995). A decline in the and from Binagady in the Transcaucasia, of relative size of P1 can be regarded as an Riss-Würm age (250.5 mm), are fully consis- index of carnivorous specialization of the tent with the European scheme. In C. spelaea dentition of the cave hyena. Such a reduction u l t i m a (MAT S U M O TO, 1915) from Ye n c h i n g - of the first premolar (up to its complete kou in China, from the Middle Pleistocene reduction in the ) is observed in diff e- (Colbert & Hooijer 1953), the length of the rent lines of Aeluroidea at the stage of transi- skull (c. 260 mm) is only slightly less than in tion to myophagy. There are no appreciable the Eurasian species of the same age, d i ffe rences in sizes of the tooth among the C. spelaea praespelaea. studied samples (Table 3).

Upper cheek teeth P 2 The large size of the tooth was typical of The length of the upper dental row P1-P4 subspecies C. s. praespelaea, C. s. ultima, changed in time analogous to the change of and particularly C. s. petralona, in which condyle basal length of the skull (Table 2). this tooth, along with other upper molars, No appreciable spatial differences are obser- was extremely wide (Kurtén & Poulianos ved. At the same time, amongst average 1977). In the samples studied, the larg e s t values of the length of the upper premolars sizes were noted for P2 from the (P2 + P3 + P4), a sample from the Geographical Society Cave, and the smallest Geographical Society Cave (90.2, n = 4-7) is sizes were noted for P2 from Lunel-Viel and prominent; it exceeds all other series known Aman-Kutan (Table 4). The teeth from for C ro c u t a from the Middle and Late L u n e l - Viel are relatively wide, whereas in Pleistocene, including Mosbach and Late Pleistocene hyenas the crown is longer.

161 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ta ble 4 Measurements (mm) of second upper premolar P2 capes of China.The ratio of length and width in Crocuta spelaea of the tooth in the samples that I have stu- died are shown in Fig. 5. It should note a relatively wide P3 from Lunel-Viel, and also an elongation of the crown in samples from the Geographical Society Cave. In the latter case the average value of P3 is maximal ( Table 5). Whereas the former parameter indicates strengthening of P3 for the crus- hing of bones during the consumption of carrion, the latter is connected with the strengthening of carnivorous tendencies during myophagy of C. spelaea. Such varia- tion could have occurred independently in d i fferent local populations. There were at least three manners of P3 modification in C . s p e l a e a: an increase of absolute (C. s. petra - It is elongated moderately lengthwise in l o n a ) and relative (C. s. intermedia) width samples from Binagady and Prolom 2 and is of the tooth and increase of the height of its elongated more strongly in samples from crown (C. s. ultima): a bone crushing adap- A r c y - s u r-C ure and Aman-Kutan. A r e l a t i v e l y tation, and also an increase of the absolute narrower P2 was noted in hyenas from the (C. s. ussurica) and relative (C. s. spelaea) Late Pleistocene of southern Siberia. length of the tooth: an adaptation to flesh c u t t i n g . The sizes of P3 in the C ro c u t a of the P 3 This tooth interacts with m1 during Late Pleistocene were weakly changing from bone-crushing (Schütt 1971) and therefore its Europe to the Altai, and only in the Russian structure can be used to assess the adaptation Far East they were much larger and relative- of hyena dentition to feeding on carrion. In ly narrower (in the subspecies C. s. ussurica C. spelaea from Europe the main tendency of Baryshnikov & Vereshchagin, 1996). changing sizes and proportions of P3 is that in animals of interglacial epochs it was shor- P 4 In the evolution of C. spelaea one obser- ter and wider than in hyenas of glacial ves an increase of total sizes of P4 and the epochs (Kurtén & Poulianos 1977). A r e l a t i - relative length of its metastyle towards the ve increase of the third premolar (as compa- Late Pleistocene (Kurtén 1956, Kurtén & red to the size of the carnassial tooth) is Poulianos 1977). My data also indicate that observed in C. s. doro t h e a e (Kurtén 1965) length of tooth varied in accordance with from the Levant. As regards C. s. ultima fluctuations of sizes, attaining larg e r from the Middle Pleistocene of China, this values in glacial epochs (Table 6). T h e subspecies is characterized by the extremely l a rgest mean value of length of the upper high crown of P3 (Kurtén 1956). This de- carnassial tooth was found in the sample gree of increase in the height of the crown from Geographical Society Cave. (hypsodontia) was not noted in other popu- lations of C. spelaea, including the collec- The relative length of the metastyle of P4 tion from the Geographical Society Cave. It (Lm/L) characterizes a specialization of indicates a specialization in C. s. ultima t o Carnivora to myophagy, and therefore it can splitting thick-walled tubular bones and be an index of carnivorous adaptation of C . tearing apart the remains of giant herbivores s p e l a e a. In the Mindel and Würm epochs the that were abundant in the Pleistocene lands- metastyle was elongating, whereas in the

162 BARYSHNIKOV: Pleistocene Crocuta from Russia

Figure 5 Scatterg ram of width (W) and length (L) of upper third premolar P3 in Crocuta spelaea. a - A r c y - s u r- C u r e , b - Binagady, d - A l t a i , g - Golabiec, j - Jaurens, k - Kazanka Rive r, l - Lunel-Viel, m - A m a n - K u t a n , p - Prolom 2, u - Geographical Society Cave

Mindel-Riss and Riss-Würm it was relatively General re m a r k s Among the upper teeth, shortened (Fig. 6). The elongation of the most important in the evolution of C ro c u t a metastyle in the glacial epochs of the are transformations of the P3 and P4. My Pleistocene indicates the growing role of analysis of the main components of length active predation in the life of the cave hyena. and width in both teeth in C. spelaea and More predatory habits were determined by C. cro c u t a ( Table 7; Fig. 7) shows that the the necessity to increase the consumption of samples used were divided in factor space F1 flesh with its high caloric value, particularly x F2 into three groups. The greatest devia- in northern populations that lived in a cold tion was noted for C. s. petralona, known climate with a long and harsh winter period. only from isolated specimens. This form was According to this character, the most advan- characterized by unusually wide upper pre- ced animals among the Würm C ro c u t a w e r e molars (Kurtén & Poulianos 1977). T h e from the northwestern part of the distribution second group consisted of recent C. cro c u t a, range (Arcy-sur-Cure). In the C ro c u t a f r o m which is confirmed by its isolation from C . the Geographical Society Cave, on the other s p e l a e a. As regards the pattern of samples of hand, the general size of the carnassial tooth C. spelaea, four groups are distinguished was larger and therefore the relative length within it: (b, d, l), (a, p, g), (m), (u). Each of of the metastyles was somewhat smaller. the last three groups corresponds to subspe- cies C. s. spelaea, C. s. praespelaea and C. s. ussurica. The first group consists of

163 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ta ble 5 Measurements (mm) of third upper premolar P3 in Ta ble 6 Measurements (mm) of fo u r th upper premolar P4 in Crocuta spelaea. Crocuta spelaea.

subspecies C. s. doro t h e a e, C. s. intermedia, smallest occurred in cave hyenas from and a sample from Binagady. It does not con- Mindel-Riss and Riss-Würm (Table 8). T h e tradict the inclusion of C. s. dorotheae i n t o sum of average indices of the length of four the species C. spelaea, and also probably lower cheek teeth (p2+p3+p4+m1) attains its suggests that C. spelaea from Binagady maximal value in materials from Mosbach deserves being placed in a separate subspe- and Petralona Cave (Schütt 1971, Kurtén & cies (together with other hyenas of the last Poulianos 1977); in the Geographical Society i n t e rgl acial in Europe). Cave it was much lower (99.8 mm, n = 8- 13). Among Würm C ro c u t a individuals from L ower cheek teeth Siberia the greatest lengths of P/2-M/1 are The distribution of average length values of f o u n d . the lower dental row p2-m1 in time shows that the largest sizes occurred in C ro c u t a p 2 The size and proportions of p2 in the from Cromerian-Mindel and Würm, and the studied material vary slightly and only in the sample from Lunel-Viel are its sizes notably smaller (Table 9). Relatively narrower were specimens from Binagady and A m a n - K u t a n Ta ble 7 Correlations of chara c t e rs with the fi r st three (as compared to samples from A r c y - s u r - p rincipal components for upper cheek teeth. Cure, Prolom 2, and the Geographical Society Cave). However, this variation is most likely not related to the sites' southern locations, since a subspecies which lived even farther south, C. s. doro t h e a e, had a relatively wide p2 (Kurtén 1965). Whereas in C. s. praespelaea the p2 was relatively nar- r o w, in C. s. petralona it had a relatively wide crown (Schütt 1971, Kurtén & Poulianos 1977).

p 3 Unlike in the P3, the sizes and propor- tions of p3 were changing slightly in the

164 BARYSHNIKOV: Pleistocene Crocuta from Russia

Figure 6 Scatterg ram of metastyle length (Lm) and maximal length (L) of fo u r th upper premolar P4 in Crocuta spelaea. Symbols as in Fig 5.

evolution of C. spelaea. In bone-crushing the Cave, Aman-Kutan, A r c y - s u r-Cure, and function of the lower premolar was evidently Golobiec, and relatively narrow in Binagady less significant than that of the upper one and the Geographical Society Cave. and it participated to a lesser degree in the process of dental specialization The small p 4 Variation in sizes of p4 had the same sizes of p3 are typical of samples from tendencies as in other lower premolars: its L u n e l - Viel, Binagady, Ilskaya 1, and A m a n - length increased in glacial times and decli- Kutan, and large sizes are typical of A r c y - ned in the interglacials (Table 11). The relati- s u r-C ure, sites of the Urals, and the ve width of the tooth reveals insignificant Geographical Society Cave (Table 10). C. s. v a r i a b i l i t y, and only in animals from p r a e s p e l a e a and C. s. petralona also posses- Mosbach, Petralona Cave, and Binagady was sed a large p3. There is no chronostratigrap- the tooth slightly narrower. The ratio of aver- hic pattern in the changes of proportions of age values of the length of the p4 and the the crown: it is relatively wide in Petralona length of m1 was approximately the same,

165 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ta ble 8 Geographical variability of length of p2-m1 (mm) in Ta ble 9 Measurements (mm) of second lower premolar p2 Crocuta spelaea in Crocuta spelaea

Whereas in the C. s. praespelaea f r o m Mosbach its length constituted on the avera- ge 12% of the total length of the tooth (Schütt 1971), in the Würm samples it did not exceed 10% (Baryshnikov 1995). T h e ratio of the length of the trigonid and the maximal length of m1 is shown in Fig. 8. The presence of a small metaconid on m1 varies considerably, and, as has been noted before (Kretzoi 1938, Kurtén 1956), is not a reliable systematic character. The frequency of occurrence of the metaconid varied from 23-29% in the Cromerian and Mindel but in Mosbach and Petralona Cave the p4 (Mosbach, Petralona Cave) to 50% in the was relatively more elongated (Schütt 1971, Mindel-Riss (Lunel-Viel), and then declined Kurtén & Poulianos 1977). to 38% in the Riss (Chatillon-Saint-Jean), increased to 55% in the Riss-Würm m 1 The length of the lower carnassial tooth (Binagady) and declined again to 41-43% in in C. spelaea also changed during the the Würm (Schütt 1971, Ballesio 1979, Pleistocene depending on the temperature Baryshnikov & Vereshchagin 1996). factor (Table 12). Apart from that one can According to the data of Kurtén & Poulianos see distinctly that the length of m1 increased (1977), the occurrence of the character on from south to north, from its minimal avera- Riss-Würm sites of Europe is very low ge values in Aman-Kutan to maximal ones in ( 11%). During the Würm the frequency of Jaurens and the Geographical Society Cave. the presence of a metaconid declined from In the evolution of C. spelaea the talonid of the Atlantic coast towards continental parts m1 became simpler and shorter (reflecting of Europe and Siberia: 46% in Britain probably a specialization to myophagy). (Kent's Cavern, n=37; Kurtén 1956), 30% in

166 BARYSHNIKOV: Pleistocene Crocuta from Russia

Figure 7 Principal component analysis based on length and width of P3 and length and width of P4 in Crocuta spelaea. a - A r c r y - s u r- C u r e , b B i n a g a d y, c - C. c r o c u t a,recent (coll. Z I N ) , d - C. s. d o r o t h e a e ( K u r t é n , 1 9 6 5 ) , e - C. s. p e t ra l o n a ( K u r tén & Po u l i a n o s , 1 9 7 7 ) , g - Golabiec j - Jaurens, k - A m a n - K u t a n , 1 - Lunel-Viel, m - C. s. p ra e s p e l a e a,Mosbach (Schütt, 1 9 7 1 ) , p - Prolom 2, u - G e o g raphical Society Cave.

France (Arcy-sur-Cure, Jaurens, n=23), 27% General re m a r k s Characterizing the in the Crimea (Prolom 2, Ay u v u l - K o b a , lower molars of C. spelaea, one should note n = 11), 25% in Western Siberia (Krasnyi Ya r, that the decrease or increase of their sizes Ust-Kanskaya Cave, Denisova Cave, n=4), was related to the change of the sizes of the 29% in the Russian Far East (Geographical animals. Unlike the upper molars they do not Society Cave, n=7). demonstrate regular changes in crown pro- portions on different temporal levels of the The above noted variation in the presence of Pleistocene. We only observe a strengthening a metaconid may be related to the temperatu- of the cutting character of m1 (a shortening re factor, the type of feeding of the animals, of the talonid) and reduction of its crushing and peculiarities of local populations, in par- structures (a decline of the occurrence fre- ticular influenced by weakening of inter- quency of a metaconid) in glacial epochs and populational exchange of genes in the case of in regions with a more rigorous climate. a more territorial or isolated way of life and Analysis of the main components for length restricted interclan contacts. and width of p3 and m1 (Table 13; Fig. 9) did not indicate a distinct pattern in the dis- tribution of samples. In the factor space F1xF2 they form seven groups (e), (c), (j), (m, u), (b), (a, g, p), (d, k, l). Of these the

167 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Figure 8 Scatterg r am of trigonid length (Lt) and maximum length (L) of fi r st lower molar m1 in Crocuta spelaea. Symbols as in Fig. 5 .

first two groups including C. s. petralona regularity known as ‘Bergman's rule’. T h a t and recent C. cro c u t a are the most remote the change was determined by the climate is from the others. It noteworthy that the samp- confirmed by an analogous variation of its le from Binagady is separated, and that the sizes observed in other Carnivora from the small C ro c u t a that inhabited a warmer clima- Pleistocene of Eurasia (Kurtén 1968). te are joined in one group: C. s. doro t h e a e, C. s. intermedia, and the hyena from A m a n - The dentition of C ro c u t a is most advanced in K u t a n . the Hyaenidae. Its transformation is of a dualistic nature: a parallel specialization to D I S C U S S I O N myophagy (elongation of P4 and particularly E vo l u t i o n a r y tre n d s of its blade-shaped metastyle, and also a size The paleontological record indicates that increase of m1, accompanied by a reduction sizes of C. spelaea increased in the glacial of the metaconid, and a shortening and sim- epochs and decreased in the interg l a c i a l plification of the talonid) and a specialization periods, and also increased from south to the crushing of large bones (an increase of towards the northeast within the same time the width and height of the crown of the P3 interval. Therefore, change in the sizes of and p3). Analogous transformations have animals depended on the temperature factor been noted also for the deciduous cheek teeth and occurred according to the zoogeographic (Baryshnikov & Averianov 1995).

168 BARYSHNIKOV: Pleistocene Crocuta from Russia

Figure 9 Principal component analysis based on width and length of p3 and width and length of m1 in Crocuta spelaea. Symbols as in Fig. 7 .

The main evolutionary tendency of C. spela - conditions, abundant food, and the consu- e a is connected with the strengthening of the ming of carrion) and a reverse shift in glacia- carnivorous specialization of its dentition. It tion epochs (unfavorable climatic conditions, was the transition to myophagy, to active the need for more higher-calory flesh, an hunting in large groups, that permitted C . increase in hunting activity, which was s p e l a e a to inhabit geographic regions with accompanied by competition for food with long cold winters. However, the development other large and primitive of the cutting function of the teeth was neither h u m a n s ) . unidirectional nor accumulative. Phases of carnivorous specialization associated with As in many other Pleistocene , such glacial periods were followed by despeciali- as S p e l e a rc t o s , for example (Baryshnikov zation phases in the interglacials. During the 1997), the rates of evolutionary change of latter a return to the previous stage and deve- the upper molars in C. spelaea were somew- lopment of bone-crushing rather than specia- hat higher than of lower ones. This is indica- lized cutting dentition occurred. Such varia- ted by the fact that the difference between tion in the dentition of C. spelaea can be hyenas of the glacial and interglacial epochs accounted for by a shift in natural selection were more distinct in the upper molars' pressure on populations of the cave hyena in m o r p h o m e t r y. the interglacials (i.e., by favorable climatic

169 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ta ble 10 Measurements (mm) of third lower premolar p3 in Ta ble 11 Measurements (mm) of fo u r th lower premolar Crocuta spelaea. p4 in Crocuta spelaea.

G e og r aphical variability and taxonomy Eurasia. These forms may be regarded as Kurtén (1956) distinguished two extinct sub- both stratigraphic and geographic subspecies species of C. cro c u t a in Eurasia: C. cro c u t a of C. spelaea. In the Middle Pleistocene the s p e l a e a (GO L D F U S S ) in Europa and C. c. ulti - following forms are distinguished: C. s. prae- m a (MAT S U M O T O) in China, thereby acknow- s p e l a e a (Mosbach, Cromerian-Mindel), C. s. ledging the geographic difference between p e t r a l o n a (Petralona Cave, Mindel), C. s. hyenas from western and eastern parts of the i n t e r m e d i a ( L u n e l - Viel, Mindel-Riss) and C . distribution range. Later, several more forms s. ultima (from a certain marly district of were described for the Pleistocene of Szechuan in China, Middle Pleistocene;

Ta ble 12 Measurements (mm) of fi r st lower molar m1 in Crocuta spelaea. Ta ble 13 Correlations of chara c t e rs with the fi rst three p rincipal components for lower cheek teeth.

170 BARYSHNIKOV: Pleistocene Crocuta from Russia

F i g . 10 Similarity dendrogram of samples of genus C r o c u t a based on length means of upper P2-P4 and lower p2-m1. M a t rix and samples are in Ta ble 14.

Matsumoto 1915, Bonifay 1971, Schütt to the Urals and possibly south of We s t e r n 1971, Kurtén & Poulianos 1977). If we igno- Siberia and Kazakhstan. A very small form re C. s. petralona, which was a strongly C. s. doro t h e a is known only from the deviating local form, the other subspecies Levant. Similar to it in teeth sizes is the supplant each other in a succession in time C ro c u t a from Aman-Kutan, but this form's and space, demonstrating different trends of taxonomic position is not clear yet. South of dental specialization. the Primorski Krai of Russia and northeas- tern China lived the large-toothed C. s. From the Late Pleistocene onwards, the sub- u s s u r i c a , the distribution range of which pro- species are based mainly on geographic bably spread westwards up to the A l t a i . variation in sizes: C. s. spelaea ( G a i l e n - reuther Cave, Würm), C. s. dorotheae ( Z u t t i y e h Two groups can be distinguished in the simi- Cave, Würm), and C. s. ussurica ( G e o g r a p h i - larity-dendrogram of C ro c u t a-samples by cal Society Cave, Würm; Goldfuss 1823, average values of the length of the upper Kurtén 1965, Baryshnikov & Ve r e s h c h a g i n (P2-P4) and lower (p2-m1) cheek teeth 1996). I refer cave hyena from the end of the ( Table 14; Fig. 10). One includes a small Middle Pleistocene (Riss) and the Riss- C ro c u t a from Africa (C. cro c u t a), two from Würm to C. s. spelaea, although it is possi- the tropical and subtropical zones (C. siva - ble that this form represents a separate sub- l e n s i s, C. s. doro t h e a e ), and three from the species. The distribution range of C. s. spela - i n t e rglacial epochs of the temperate zone of e a covered the whole of Europe eastwards up Eurasia (C. s. intermedia, C. s. spelaea, the

171 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

Ta ble 14 Matrix of norm a l i zed euclidean distances within the 12 samples of C r o c u t a based on mean measurements of length of upper (P2-P4) and lower (p2-m1) teeth; K e y : 1 = C. s i va l e n s i s ( K u r tén 1956); 2 = C. c r o c u t a ( K u r tén 1956); 3 = C. s. p ra e s p e l a e a, Mosbach (Schütt 1971); 4 = C. s. p e t ra l o n a ( K u r tén & Poulianos 1977); 5 = C. s. u l t i m a ( K u r tén 1956); 6 = C. s. i n t e r - m e d i a, L u n e l - V i e l ; 7 = C. s. s p e l a e a, B i n a g a d y ; 8 = C. s. s p e l a e a,Kent's Cave rn (Kurtén 1956); 9 = C. s. s p e l a e a,A r c y - s u r- C u r e ; 1 0 = C. s. s p e l a e a,Prolom 2; 1 1 = C. s. u s s u ri c a, G e o g raphical Society Cave ; 1 2 = C. s. d o r o t h e a e ( K u r tén 1965).

Binagady form). The other group includes AC K N OW L E D G E M E N T S two subgroups: a) large hyenas of cold I would like to thank Prof. N. Ve r e s h c h a g i n epochs from the European Late Pleistocene (St. Petersburg, Russia), Dr. N. Fomicheva (C. s. spelaea, different localities) and from (Kazan, Russia), Prof. A. Derevianko the Middle Pleistocene of China (C. s. (Novosibirsk, Russia), Prof. Yu. Kolossov u l t i m a); b) very large hyenas from the ( K i e v, Ukraine), Dr. A. Nadachowski Middle Pleistocene of Europe (C. s. prae - ( K r a k o w, Poland), Dr. C. Mourer- C h a u v i r é s p e l a e a, C. s. petralona) and from the Late ( Lyon, France), Dr. F. David (Paris, France), Pleistocene of the Ussuri River region (C. s. Prof. M.-F. Bonifay and Dr. J.-Ph. Brugal u s s u r i c a ). Note should be made of the isola- (Marseille, France) for their help in studying ted position of C. s. ussurica in the dendro- collections. I thank Dr. A. Averianov (St. gram, which confirms the hypothesis that P e t e r s b u rg) and r e v i ewer Prof. R. Klein this geographic form is a separate subspe- (Stanford University, USA) for his comments c i e s . on the manuscript. The manuscript was trans- lated into English by Tatyana Platonova and Olga Potapova, edited by Prof. Gary Haynes (Reno, USA).

172 BARYSHNIKOV: Pleistocene Crocuta from Russia

R E F E R E N C E S Goldfuss, A., 1823 - Osteologische Beiträge zur Ballesio, R., 1979 - Le gisement Pleistocène supérieur Kenntnis verschiedener Saugthiere der Vorwelt - de la grotte de Jaurens à Nespouls, Corrèze, France: Nova Acta Physico-Medica Academiae Cesareae les Carnivores (Mammalia, Carnivora). I. Canidae et Leopoldino-Carolinae 21: 449-490 Hyaenidae - Nouvelles Archives du Muséum Klein, R., & Scott, K., 1989 - Glacial/Interglacial size d'Histoire naturelle de Lyon 17: 25-55 variation in fossil spotted hyenas (C rocuta cro c u t a) B a r y s h n i k o v, G., 1995 - Cave hyena, C rocuta spelaea from Britain - Quaternary Research 32: 88-95 (Carnivora, Hyaenidae) from Palaeolithic fauna of Kretzoi, M., 1938 - Die Raubtiere von Gombaszög nebst the Crimea - Trudy Zoologicheskogo Instituta RAN einer Übersicht der Gesamtfauna - Annales Museum 263: 3-45 Nationalis Hungaricum. Pars Mineralogica, B a r y s h n i k o v, G., 1997 - Cave bears from the Palaeo- Geologica, Palaeontologica 31: 88-157 lithic of the Greater Caucasus - in Saunders, J.J. Kurtén, B., 1956 - The status and affinities of H y a e n a Styles, B.W. & Baryshnikov, G.F. (eds.) s i n e n s i s Owen and Hyaena ultima Matsumoto - Quaternary Paleozoology in the Northern American Museum Novitates 1764: 1-48 Hemisphere - Illinois State Museum Scientific Kurtén, B., 1957 - The bears and hyaenas of the Papers 27: 45-97. Springfield i n t e rgl acials - Quaternaria 4: 69-81 B a r y s h n i k o v, G. & Av e r i a n o v, A., 1995 - Deciduous Kurtén, B., 1965 - The Carnivora of the Palestine caves teeth of carnivorous mammals (order Carnivora). - Acta Zoologica Fennica 107: 1-74 Part V. Families Protelidae and Hyaenidae - Trudy Kurtén, B., 1968 - Pleistocene mammals of Europe - Zoologicheskogo Instituta RAN 263: 46-84 Aldine Publishing Co., Chicago: 317 pp. B a r y s h n i k o v, G. & Vereshchagin, N., 1996 - A brief Kurtén, B. & Poulianos, N., 1977 - New stratigraphic review of Quaternary hyenas (Hyaenidae) of Russia and faunal material from Petralona Cave with special and adjoining regions (in Russian with English reference to the Carnivora - Anthropos (Athens) 4: summary) - Trudy Zoologicheskogo Instituta RAN 47-130 270: 7-65 Lu Youquan, Li Yi & Changzhu, J., 1986 - Mammalian B o n i f a y, M.-F., 1971 - Carnivores quaternaires du remains from the Late Pleistocene of Wurji, Nei sud-est de la France - Mémoires du Muséum Mongol (in Chinese with English summary) - National d'Histoire Naturelle, Ser.C 21: 43-377 Vertebrata PalAsiatica 24: 152-162 Brugal, J.-P., Fosse, P., & Guadelli, J.-L., 1997 - Matsumoto, H., 1915 - On some fossil mammals from Comparative study of bone assemblages made by Sze-chuan, China (in Chinese) - Science Reports of Recent and Pleistocene hyenids - in: Hannus, A., the Tohoku Imperial University. Ser.2 (Geology) 3: Rossum, L., & Winham, P. (eds.) - Proceedings 1-28 of the 1993 Bone Modification Conference – Pei, W.C., 1940 - The Upper Cave Fauna of pp. 158-187. Hot Springs, South Dakota Choukoutien - Palaeontologia Sinica, Ser.C 10: 1-84. Clot, A., 1980 - La grotte de la Carrière (Gerde, Ht. Schütt, G., 1971 - Die Hyänen der Mosbacher Sande Pyrenees). Stratigraphie et paléontologie des (Altpleistozän, Wiesbaden/Hessen) mit einem carnivores - Laboratoire de Géologie, To u l o u s e Beitrag zur Stammesgeschichte der Gattung C ro c u t a Colbert, E.H. & Hooijer, D.A., 1953 - Pleistocene - Mainzer Naturwissenschaftliches Archiv 10: 29-76 mammals from the Limestone fissures of Szechwan, Werdelin, L. & Solounias, N., 1991 - The Hyaenidae: China - Bulletin of the American Museum of Natural t a x o n o m y, systematics and evolution - Fossils and History 102: 1-134 Strata 30: 1-104 Coloman, C., 1978 - C rocuta spelaea (GO L D F U S S) in pleistocenul superior de la Ripa (bazinul Beius) - Nymphaea 6: 183-204 David, A.I., 1980 - Teriofauna from the Pleistocene of Moldavia (in Russian) - Shtiintsa, Kishinev: 187 pp.

173 MAMMOTHS AND THE MAMMOTH FAUNA DEINSEA 6, 1999

DEINSEA - ANNUAL OF THE NATURAL HISTORY MUSEUM ROTTERDAM P. O . B o x 2 3 4 5 2 , N L - 3 0 0 1 K L R o t t e r d a m T h e N e t h e r l a n d s

174