GEOLOGIA CROATICA 45 l - 14 5 Fig. 2 Pl. 2 Tab. ZAGREB 1992 UDC 551.791:569.6 (497.13) Scientific paper Revision of the Ungulate Fauna and Upper Pleistocene Stratigraphy of Veternica Cave (Zagreb, Croatia) l?reston MIRACLEl and Dejana BRAJKOVJć2 Key words: paleontolo gy, ungulates, Upper Pleistocene, paleolithic, Croatia Revision of the Veternica ungulates removes Dama dama (L.) (fallow deer) and Megaceros giganteus (BLUMENBACH) (giant deer) from the species list and adds Bison priscus (BOJ ANUS) (bison) to it. The rhino remains do not allow a specific determination. A/ces a/ces (L.)(elk) and Capreolus capreolus (L.) (roe deer) are rarer than previously thought. Rupicapra rupicapra (L.) (chamois) re­ mains are particularly common in levels d and h. Capra ibex (L.) (ibex) is limited to level d, while Sus scrofa (L.)(pig) needs to be added to the faunallist of level e. Cervus ela ph us (L. )(red deer) is the most common ungulate in all of the levels, with the exception of level d and perhaps level h. This revision remove s all of the so­ called warm period ungulates from the fauna and casts doubt on an interglacial date for level j. Therefore, we suggest that level j at Veternica does not date to the last interglacial maximum (oxygen isotope sub-stage Se, e. 128-118,000 BP), but to a later and cooler time period with a more open environment. The sediments, how­ ever, do indicate relatively warm and wet conditions of deposition, at least with regards to the rest of the profile. We favor a date dur­ ing sub-stages Sc or Sa (e. 100,000 and 80,000 BP, respectively), but leave the question open for the time being. l. INTRODUCTION Veternica cave is a well-known site of paleontological, archaeological, and speleological interest. lt is particularly important for its abundant evidence of hominid and cave bear occupation during the Upper Pleistocene period. Hominid occupation is shown by Mousterian artifacts, probably manufactured by Neandertals, and hearths, while cave bear use of the cave is inferred from the abun­ dance of th6r remains. The prehistoric significance of Veternica was recognized by the late Dr. Mirko MALEZ, wno comprehensively presented the results of his research in an admirable mono graph (MALEZ, · 1963) and several articles (MALEZ,1958/59, 1965). A paleontological revision of the Veternica ungu­ lates was undertaken as part of a larger study by the senior author of the taphonomy of cave sites in Croatia, and the evolution of hominid economies in this region during the Upper Pleistocene. In a recent publication, one of us (MIRACLE, 1991) suggested that many of the cave faunas from middle and early upper paleolithic sites may bear a very weak inmrint of hominid activi­ ties. In particular, it was hypothesized, thatassemblages Ključne riječi: paleontologija, ungulati, gornji pleistocen, paleolit, Hrvatska Revizija fosilne građe skupine ungulata Veternice pokazala je odsutnost vrsta Dama dama (L.) iMegaceros gig ante us (BLUMENBACH) i prisutnost vrste Bison priscus (BOJ ANUS) . Ostaci nosoroga nisu se pokazali dostatnim za odredbu vrste. U ukupnom fosilnom materijalu skupine ungulata, vrste A/ces a/ces (L.) i Capreolus capreolus (L.) manje su zastupljene nego što se do sada mislilo dok je Rupicapra rupica pra osobito prisutna u slojevima d i h. Capra ibex (L. )ograničena je samo na sloj d, a faunskoj zajednici sloja e priključena je i vrsta Sus scrofa (L) U svim slojevima, osim d i možda h, najbrojniji su ostaci vrste Cervus elaphus L. Učinjena revizija fosilne građe skupine ungulata Vetemice ne potvrđuje svoju pripadnost toplodobnoj faunskoj zajednici kako se to do sada smatralo. Mišljenja smo da se ni sloj j ne može datirati u posljednji interglacijalni maksimum (izotop kisika stadij Se, cca 128-118.000 BP), već u kasniji i hladniji period s otvoreni jim biotopom. I u pogledu ostalih dijelova profila, sedimenti ukazuju na relativno tople i vlažne uvjete taloženja. Prednost je dana datiranjima izotopom kisika, stadijima Sc ili Sa (cca 100 ili 80.000 BP), koji bi mogli odgovarati sloju j, no to pitanje ostaje otvoreno. dominated by cave bear and large carnivore remains were accumulated primarily through non-hominid processes. While the presence of many carnivores in cave faunas may appear to be explained by the activi­ ties of carnivores, the taphonomic status of ungulates in assemblages is impossible to untangle in the absence of detailed studies of the remains themselves (MIRACLE, 1991). Remains from Veternica were originally examined as comparative material to aid the identification of remains from Šandalja II and other sites. When it became ap­ parent that there were a number of problematic iden­ tifications in the monograph (MALEZ,1963), it was decided to review and revise the ungulate remains in earnest. 2. SUMMARY OF THE VETERNICA SEQUENCE Veternica is located about 9 km west of the center of Zagreb, Croatia, on the southwestern stope of Medvednica (45°50'36" North, 13°32'24" East, 306 m above sea level, MALEZ,1965, 176). The entrance to the cave faces south southeast and is about 8 m wide and 4 m high. The entrance chamber is about 15 ·m l Museum of Anthropology; University df Michigan; Ann Arbor, MI 48109 USA 2Zavod za paleontologiju i geologiju kvartara; A. Kovačića S; 41000 Zagreb Croatia - 2 long and 7 m wide (oriented east northeast-west south west with the entrance on the southeast corner) and receives direct sunlight for most of the day. Beginning from the northwest corner of the entrance chamber, the "left hall" extends for 14 m and is J-7 m wide. Leading off of the northeast corner of the entrance chamber is a narrow passageway that gives access to over lO km of pas­ sageways and canals. · Quatemary sediments were excavated by MALEZ (1965, 178; 1972) in the entrance chamber and left hall from 1951-1955, and in 1970, over an area of ap­ proximately 207 square meters (43 m2 in front of the entrance, 84 m2 in the entrance chamber, and 80 m2 in the left hall). In front of the cave, sediments were excavated to a depth of 8 m, while the entrance chamber and left hall were excavated to bedrock. In the most complete profiles, ll geological lay­ ers were identified. The first two layers, layers a and b, were black to brown in color, rich in organic material (particularly level a), and contained archaeological and faunal remains from the N eolithic to Middle Ages. These soils formed during the Holocene and were found in front of the cave and in most parts of the entrance chamber and left hall. Within the cave, but not in front of the entrance, these soils are underlain by dripstone, level e, which was barren of any remains except for a few floral remains of the following taxa: Polypodium, Pteridium, Carex, and Corylus (MALEZ,l965, 202). Carbonate samples from the upper and middle parts of level e gave mid-Holocene C14 dates (5800±80 BP, Z- 218/11; 5960±90 BP, Z-194), while carbonate samples from the lower part of level e gave late Pleistocene C14 dates (12,400±150 BP, Z-218/1; 13,650±75 BP, GrN- 4989; 13,660±180 BP, Z-201) (MALEZ,l979a, 218). Beneath this dripstone cap, there were a series of lay­ ers deposited during the Pleistocene. Layers d, e, and f contain almost exclusively cave bear (Ursus spelaeus ) remains. Evidence of human occupation is very scan t; there are a few lithics and a "hearth" from level d (MALEZ,l965, 203), and there are several non-diag­ nostic lithics from level/ (MALEZ,l979b, 269). Level g was found primarily in front of the cave and in the entrance chamber, and was sterile of archaeological or paleontological remains excepting a few cave bear bones near the contact with level h. Level g consisted pri­ marily of rock rubble and probably marks a major roof collapse that closed the entrance to Veternica (MALEZ, 1965, 207). Layers h, i, and j contain about 178lithics that have been attributed to human handicraft, of which at least 18 are Mousterian "tools" (lithic totals calcu­ lated from MALEZ,l958/59, 173, Figure l; tool count from MALEZ,l979b, Plates XXIX and XXX). A hominid calvarium from level h, considered by MALEZ (1979b, 268) to be a neanthropic representative of "Homo sa­ piens fossilis," has reen described by others as com­ pletely modem in ma:-phology and most likely redeposited from younger levels(SMITH,l982, 682). Hearths were also found in all three levels, and charcoal from a hearth Geologia Croatica 45 in level i was radiocarbon dated at >43,200 BP (GrN- 4984) (MALEZ,l979a, 218). Artifacts and ungulate remains were found piimarily in the entrance chamber. Cave bear remains are still very common, and were found in the entrance chamber and left hall. Hearths were found in both the entrance chamber and in front of the cave. O~ the basis of the sediments, associated fau­ nas and artifacts, layers h, i, and j were dated as Wlirm l/ll interstadial, Wlirm I stadial, and Riss/Wlirm in­ terglacial respectively (MALEZ,l963, 153; 1965, 220- 222). The lowest level, k, was only found in front of the entrance and in the entrance, and did not contain any paleontological or archaeological remains. 3. PALEONTOLOGICAL REVISION Perissodactyla Dicerorhinus kirchbergensis (JAEGER) Four specimens were attributed to Merck's rhino; a left M2 (VTR. 199), a first phalanx (VTR. 200), a third phalanx (VTR. 201) and a rib shaft fragment (VTR. 202), all from levelj (MALEZ,l963, 108-112; Plate XXII; Plate XXIV, Figure-:la-c). The relatively smooth enamel of the left second molar leaves little doubt that it is from the genus Dieerorhinus. The remaining elements are very likely from the same genus, if not the same indi­ vidual as the mo lar. While metric comparisons (MALEZ, 1963, 109-110) suggest that the molar and the phalanxes are too large to be from Dieerorhinus etruscus, there is nothing to rule out the possibility that they are from Dicerorhinus hemitoechus. As observed by several authors (LOOSE,l975, 6; GUERIN,l980, 654) specific determination is extremely difficult on isolated teeth (particularly fower mo lars), let alone phalanxes and rib fragments. We suggest that a more prudent identifi­ cation of these remains is Dicerorhinus sp. Artiodactyla Megaceros giganteus (BLUMENBACH) MALEZ (1963, 114-115, Plate XXV, Figure la-d), assigned three specimens to giant deer ( VTR. l, 2, 3). These are a navicular-cuboid and two third phalanxes. The navicular-cuboid (VTR. 3,levelj) is clearly bovid in morphology. In Plate I it is compared to a navicular­ cuboid from a giant deer (see also MALEZ,1963, Plate XXV, Figure la-d). In general shape, it is broader (medio­ laterally) and shorter (superior-inferiorly) than that expected from a cervid. The superior juncture between the facets with the calcaneus and astragalus is flat, as in bovids. Similarly, the styloid on the plantar edge of the navicular is constricted medio-laterally, and in di stal view, the pos teri or articular facet with the metatarsal is ovoid in shape; both morphologies are characteris­ tic of bovids. This element is classified as Bos!Bison, since it does not show clear characteristics for specific determination. The third phalanxes (VTR. l and 2, both level i) both lack a large nutrient foramen immediately inferior and Miracle & Brajković : Revision of the Ungulate ... medial of the articular surface. A large foramen in this position is characteristic of giant deer and cervids in general. The phalanxes are also marked by well-de­ veloped articular facets for distal sesamoids on the posterior edge of the articular surface. Such facets are typical of bovids, and rare or non-existent in cervids (Plate I, Figure 2A-B). Therefore, these phalanxes are clearly from a large bovid, and are here only identified as Bos!Bison, since identification to species is very difficult (SALA, 1986, 162). Measurements made on these specimens did not differ significantly from those published by MALEZ (1963, 114). Dama cf. dama (L.) MALEZ (1963, 115, Plate XXVII, Figure 3) from level j identified an upper right M3 as fal low deer (VTR. 41). Close examination of the tooth shows a slight wear facet on its distal surface. In this case, the tooth can­ not be an M3, but rather must be either an M2, Ml or dp4. Measured at the base of the crown, its dimensions are: length (in mm) = 20.1 width (in mm) = 21.7 these dimensions are not significantly different from those published by MALEZ (1963, 115). When com­ pared to the dimensions of upper molariform red deer teeth from Upper Pleistocene levels in Sandalja II, the specimen from Veternica falls in the range of the Ml (Fig. 1). 28 i 26 :§. -5 24 "O l! ~ 22 20 18 16 o •• • Upper Teeth: Length against Breadth o " o o VTR.41 o 17 18 19 20 21 22 23 24 25 26 27 28 Length (iit mm) O Sa: Upper Ml O Sa: Upper M2 l> Ša: UpperM3 e Sa: Upperdp4 Fi g. l. Length against breadth al crown base of upper teeth in CervuS elaphus from Upper Pleislocene levels from Šandalja II (MIRACLE, un published) compared lo the molar identified as DaTNJ dama from Veternica (MALEZ, 1963, 115, Plate XXVII, Figure 3). Slika l. Usporedba dužina i širina mjerenih na bazi kruna _gornjih zuba kod vrste Cervus elaphus iz gomjopleistocenskih naslaga Sandalje II (MIRACLE, rukopis) i molara iz Veternice determiniranih kao vrsta DaTNJ daTNJ (MALEZ, 1963, 115, Tabla XXVII, slika 3). Given the morphology and dimensions of this tooth, it is most likely a right Ml from Cervus elaphus. One concludes that there are no identified remains of fal­ low deer from Veternica. A/ces a/ces (L.) ~ /, MALEZ (1963, 126-128; Plate XXV, Figure 2; Plate XXVIII, Figures l, 7, 9; Plate XXX, Figure l; Plate XXXV, Figures l and 2) identified the following specimens as elk: an upper Ml (VTR. 47, level h), a lower Mt (VTR. 46, level h), a cervical vertebra (VTR. 3 48, level i ), an unfused distal epiphysis of a metacar­ pal (VTR. 45, level h or i), a calcaneus (VTR. 44, level i ), a navicular-cuboid (VTR. 4, level h ), a first pha­ lanx (VTR. 24, level i ), two second phalanxes (both level i), and a third phalanx (VTR. 5, level h). Of these specimens, we are in agreement with the identifications of the teeth. The cervical vertebra (VTR. 48) is complete with a fusing caudal centrum and a fused cranial centrum. ·Comparison with cervical vertebra of red deer, how­ ever, shows the specimen in question to be only slightly larger in size. The state of fusion of the centra makes it unlikely that the vertebra would have been much larger in an adult. If one considers that red deer during the Pleistocene were· often larger than recent examples, then it appears quite likely that the specimen in question is from a red deer. In the absence of more complete comparative material, however, identification beyond the level of cervid is not warranted. The unfu~~d di­ stal epiphysis of a metacarpal (VTR. 45) is much closer in size to red deer than elk, and in fact, falls within the range of variation of other material identified as red deer from the site. The surface of the bone is compact and non-porous, suggesting that the epiphysis would not have grown much larger. Therefore, we also identify this specimen as from red deer. The navicular-cuboid (VTR. 4) is extremely small for elk, and as shown in Figure 2, better fits into the range of red deer; although the specimen is certainly from a very large individual. Navlcular-cubold: GB against GD 60r-~~~--~~~~~--~~ 57.5 55 52.5 ~ 50 :§.47.5 o Cl 45 42.5 40 :• 37.5 • o o ..• e\ • VTR.4 42.5 45 47.5 50 52.5 55 57.5 60 62.5 65 67.5 GB(inmm) o A/ces e Cervus Fig. 2. Navicular-cuboid, greatest medio-lateral breadth (GB) against greatest dorsal-palmar depth (GD) in Cervus and A/ces, compared lo the navicu1ar-cuboid identified as A/ces from Veternica (MALEZ, 1963, 127, Plate XXV, Figure 2). Comparative data for red deer are from Sandalj a II (MIRACLE, unpublished) and Veternica. Comparative data for elk are from Pleistocene (MALEZ, 1963, 127) and recent localities (MALEZ, 1963; MIRACLE, unpublished). Slika 2. Naviculare-cuboideum, odnos najveće širine (GB) i najveće udaljenosti (dorsalno-palmamo, GD) rodova Cervus i A/ces u usporedbi s navikulamokuboidnom kosti roda A/ces iz V eternice (MALEZ, 1963, 127, Tabla XXV, slika 2). Usporedni podaci za jelena su izŠandalje II (MIRACLE, rukopis) i Veternice. Usporedni po.daci za losa su iz pleistocenskih (MALEZ, 1963, 127) i recentnih lokaliteta (MALEZ, 1963; MIRACLE, rukopis). The calcaneus (VTR. 44) is in morphology from a red deer, although from a very large individual {Fig­ ure 3). In red deer, the plantar edge ofthe articulation - 4 with the navicular-cuboid is extended medially and fonns a right angle with the body of the sustentaculum tali. In elk, this articulation is not extended medially and fe) u odnosu na širinu proksimalnog kraja (BP) rodova Cervus i A/ces u odnosu na prvu falangu roda Alas iz Vetemice (MALEZ,l963, 127, Tabla XXXV, slike 1-2). Usporedni podaci za jelena su iz Šandalje Il (MIRACLE,rukopis) i Vetemice. Usporedni podaci za losa su iz pleistocenskih (MALEZ,l963, 127) i-recentnih l~aliteta (MALEZ, 1963; MIRACLE,rukopis). - - . . Our measurements differ somewhat from those published by MALEZ (1963, 127). The measures are Geologia Croatica 45 as follows (in mm): Glpe(greatest length of peripheral) = 65.9 BP (breadth of proximal) = 25.3 Measures were not taken on the distal end due to a light abrasion of the distal condyles on their palmar surface. We also classify this specimen as red deer. All second phalanxes of cervids were locat~. but it was not possible to separate between those previously identified as red deer· and elk (the material was mixed during curation, and none of the measurements correspond to those published by MALEZ [1963, 128] as elk). All of these specimens are well within the expected size range for red deer, and we have thus identified all of them as red deer. The third phalanx (VTR. 5) has been abraded on the anterior-medial surface. It lacks a large nutrient foramen imltlediately inferior and medial of the articular surface, has a well­ developed articular facet for a distal sesamoid, and the articular surface with the second phalanx eJS.tends supe­ rior-posteriorly in a manner characteristic of bovi ds. As discussed above for the giant deer, these are characteris­ tics of bovids, and we thus classify the specimen as B os/ Bison. Only two of the ten specimens originally identi­ fied as elk appear to be elk after further scrutiny. Cervidae inc. sed. The right proximal metatarsal (VTR. 43, level b) is in morphology clearly from a bovid. Its dimensions (in mm) are as follows: BP (breadth proximal) = 44.5 Considering its small size it is identified as Bos taurus. The vertebra (VTR. 118, level f) is a thoracic (not a cervical as previously identified), and is most probably from a red deer; its identification has not been changed. The mandible fragment without teeth (VTR. 125, level unknown) is from a bovid, and thus is classified as Bosi Bison. The left dp4 (VTR. 127, level i ) is in mor­ phology and size from a red deer. Capreolus capreolus (L.) MALEZ (1963, 129-130; Plate XXXVI, Figures l, 4-6; Plate XL, Figure 4) identified the following specimens as roe deer: a cast right antler (VTR. 148, level e), a fragment of right maxilla with dp3 and dp4 (VTR. 149, level d), a fragment of left mandible with P4, M1, Mz (VTR. 150, level i ), a fragment of left mandible with dp4 and M1 (VTR. 151, level d), a right scapula (VTR. 152, level h), a left innominate (VTR. 153, level h), a left distal humerus (VTR. 154, level h), a right femur (VTR. 157, level d), a right dis tal tibia (VTR. 156, level j), a left calcaneus (VTR. 158, level e ), a right astragalus (VTR. 159, level e), a right 2nd-3rd carpal (VTR. 160), and a second phalanx (VTR. 155, level h). The right maxilla (MALEZ,l963, Plate XXXVI, Fig­ ure 5) contains permanent teeth, a p3 and p4_ The right scapula is caprid in morphology; the glenoid fossa is sub-rectangular in shape and the supra-glenoid tubercle projects inferiorly of the articular plane. Its dimensions Miracle & Brajković : Revision of the Ungulate ... are as follows (in mm): length (anterior-posterior) of proximal end = 32.6 length (anterior-posterior) of glenoid cavity = 25.2 breadth (medial-lateral) of glenoid cavity = 20.7 In morphology and size this scapula correspon~to that from the chamois. On the femur both proximal and distal epiphyses are unfused, making identification difficult Nonetheless, the distal shaft has a lip on the an teri or edge that is characteristic of caprids, and thus we identify it as either chamois or ibex. The calcaneus (MALEZ,1963, 130; Plate XL, Figure 4a-b) shows clear caprid morphology, especially in the extension of the astragalar facet inferiorly to the point that it contacts the articular facet for the navicular -cuboid. Again, we identify it as chamois. Finally, the second phalanx shows clear caprid morphology on both its proximal and distal articular surfaces. Due to its relatively small size and gracile morphology, we identify it as chamois. These changes neither remove nor add roe deer to the faunallists from any of the levels. Its remains, however, are much less common than previously thought, particularly in level h. Bos primigenius (BOJ ANUS) MALEZ (1963, 130-133; Plate XXXVII, Plate XXXVIII; Plate XXXIX, Figures 2, 3) identified the following specimens as aurochs: a left Mt (VTR. 196, level j), a left Mz (VTR. 197, level j), a left dp4 (VTR. 181, level j), a fragment of left dentary without teeth (VTR. 187, level j ), two thoracic vertebra (VTR. 182 and 198, both level h), a left dis tal humerus (VTR. 186, level j), a left diaphysis of a radius (VTR. 188, level ifj), three left metacarpal fragments (VTR. 183 and 184, level i; VTR. 185, level i/;), and one first phalanx (VTR. 187, level j). He specifically identified the humerus and metacarpal fragments as Bos; the remainder of the material was assigned to Bos on the basis of its asso­ ciation with these elements (MALEZ,1963, 131). Reconsideration of the humeral and metacarpal fragments suggests that they are either from Bison or do not allow specific determination. To identify the distal humerus, one orients the bone so that the shaft is perpendicular to a horizontal plane. In this orienta­ tion, the medial epicot:~dyle (epitrochlea) of Bos is extended distally so that it forms the most distal part of the humerus, while in Bison, the trochlea and me­ dial epicondyle are on the same horizontal plane, i.e. the medial epicondyle does not extend distally below the tr~hlea (SALA 1986, 138). The distal humerus (VTR. 186) is slightly damaged on the epitrochlea, making specific identification difficult. Nonetheless, the damage is very superficial; it is unlikely that the medial epicondyle would have extended distally below the trochlea (see MALEZ,1963, Plate XXXVIII, Figure la­ b, Plate XXXIX, Figure 3). This specimen is most likely from Bison, probably Bison priscus. Its dimensions are as follows (in mm, the latter two measurements after LEHMANN, 1949): Breadth of distal end = Breadth of trochlea = Depth of capitulum = Depth of trochlea = 108.7 105.7 46.8 64.9 5 In identifying distal metacarpal fragments, among other criteria, in Bos, the suture between the diaphysis and epiphysis is not evident in adults, while in Biso n, the position of the s uture is visible and at this point the bone is widened (SALA,1986, 143-144). One distal metacarpal fragment, VTR. 183, a lateral portion of a left metacarpal, shows a very distinct swelling of the bone at the position of the s uture (Plate II, Figure lA). Also, comparison with Bos material from Sandalja II (Plate II, Figure lB), and Biso n material from Vindija shows similarities with Biso n. Therefore, this specimen is classified as Bison. The other distal metacarpal fragment (VTR. 185) is a medial epicondyle of a left metacarpal. The bone is broken on or below the su­ ture, and it is impossible to identify it to species. It is classified as Bos!Bison. The remaining left proximal fragment (VTR. 184) includes the media! articular surface, which has a rounded dorso-medial angle and a slightly quadrangular or trapezoidal shape. On the basis of its morphology alone, one cannot specifically identify it. However, if it is part of the same metacarpal as VTR. 183, which seems likely, then it is of Bison, and this is how we classify it. Both molars, the deciduous premolar, and the dentary fragment are all part of the same left den tary. As MALEZ (1963, 131) notes, it is difficult to specifically iden­ tify isolated teeth. Nonetheless, both of the molars display features reminiscent of Biso n. The left Mt (VTR. 196) has a narrow ed neck and a pronounced swelling of the enamel at the base of the crown (Plate II, Figure 2A-B). Furthermore, it appears compressed mesio-distally (see MALEZ,l963, Plate XXXVII, Figure 1). All.ofthese features are typical of Bison (SALA, 1986, 120). The left Mz (VTR. 197, see MALEZ,1963, Plate XXXVII, Figure 2) is broken at the base of the crown. Never­ theless, there is a pronounced swelling of the enamel at the crown base. On the basis of the morphologies outlined above, both teeth are identified as B iso n priscus. The left dp4 (VTR. 181, see MALEZ, 1963, Plate XXVIII, Figure 2a-b) also shows an enamel swelling. at the base of the crown. This left den tary with three teeth is from a relatively young individual, probably between oneand two years of age since the Mz is just coming into wear (eruption time pf the Mz-in Bison bison is about 18 months, Prisan and Reher 1970 cited in GIFFORD.­ GONZALEZ, 1990, 58, eruption time in Biso n bo nasus is between 12 and24 months, WEGRZYN & SERW ATKA, 1984, 118). Specific identification based-OO-the material itself is very difficult. We place it in Bison priscus because Bison schoetensacki is found only in localities from the Middle Pleistocene (older than levels i and j in Veternica), while Bison bonasus is restricted to locali- - 6 ties from the Hol ocene (younger than levels i . and j in Veternica). We further note that m Bison bonasus, the breadth of the trochlea in the di stu humerus ranges from 74.5 mm - 98.6 mm (LEHMANN,l949, Table 181), considerably smaller than the specim~n from Veternica (105.7 mm). We are most likely dealing with the "steppe" bison, Bison priscus. The remaining elements lend themselves neither to generic nor specific identification. Therefore, the thoracic vertebra {VTR. 182 and 198), a left diaphysis of a ra­ dius {VTR. 188), and the first phalanx {VTR. 187) are classified as from Bos/Bison. We note that the first phalanx is from the anterior limb. To them we can add the right patella already classified asBos!Bison (MALEZ, 1963, 133). Rupicapra rupicapra (L.) We suggest only one change to the remains identi­ fied as chamois {MALEZ, 1963, 133-134 ); the proximal tibia {VTR. 164) from level e is in morphology from a roe deer. Capra ibex (L.) We have only one change for the ibex material (MALEZ,1963, 134-136); the right innominate {VTR. 165, level e) identified as ibex (MALEZ,1963, 135) is in morphology clearly from a pig. Capra sp. All of the remains classified as from an undetermined caprid (MALEZ,1963, 136-137) easily fall within the Inventory Geologia Croatica 45 range of variation of chamois. This adds to the cham­ ois inventory from level h a radius (VTR. 169), metatarsal (VTR. 170), and lst phalanx (VTR. 171). Ovis sp. The right humerus (VTR. 172, level d ) identified as an undetermined sheep (MALEZ,l963, 137) is in morphology and by dimensions most probably cham­ ois. Its dimensions (in mm) are: BD (breadth of distal end) = 32.8 The results of this revision are summarized in Tables l and 2. Fallow and giant deer have been removed from the species list, while bison has been added. Elk and roe deer are rarer than previously thought, while ll specimens are now identif~ed as chamois. Ch~mois remains are particularly common in levels d and h. lbex remains are found only in level d, while pig needs to be added to the faunallist of level e. On the other hand, red deer remains are slightly more common than pre­ viously thought, and include specimens from relatively large individuals. Red deer remains the most common ungulate in all of the levels, with the ex~eption of level d and perhaps level h (if none of the h/il j material is from level h ). 4. DISCUSSION The implications of this revision for the paleoeco­ logical and stratigraphic interpretation of Veternica, particularly level j, are significant. Level j, the low- Identification Numbers MALEZ 1963 Present study VTR. 1-3 Megaceros giganteus Bos/Bison VTR. 5 A/ces a/ces Bos/Bison VTR. 4, 24, 44, 45 A/ces alces Cervus elaphus VTR. 41 Dama dama Cervus elaphus VTR.43 · Cervus sp. Bos laurus VTR.48 A/ces a/ces Cervidae VTR. 125 Cervus sp. Bos/Bison VTR. 127 Cervus sp. Cervus elaphus VTR. 152, 155, 158 Capreolus capreolus Rupicapra rupicapra VTR. 157 Capreolus capreolus Rupicapra/Capra VTR. 164 Rupicapra rupicapra Capreolus capreolus VTR. 165 Capra ibex Sus scrofa VTR. 169-171 Capra sp. Rupicapra rupicapra VTR. 172 Ovis sp. Rupicapra rupicapra VTR. 181, 183, 184, Bos primigenius Bison priscus 186, 187, 196, 197 VTR. 182, 185, 187 Bos primigenius . Bos/Bison 188,189 VTR. !99-202 Dicerorhinus kirchbergensis Dicerorhinus sp. 2 Phalanx II A/ces a/ces Cervus elaphus Table l. Summary of paleontological revisions to the identified Veternica ungulate assemblage. Tablica l. Rezultati paleontololke revizije skupine ungulata iz spilje Vetemice Miracle & Brajković : Revision of the Ungulate ... 7 NISP by level Taxon d e f h l i/ p h i j Total Dicerorhinus sp . 4 4 Sus scrofa 4 l l 6 A/ces a/ces 2 2 Cervus elaphus 4 5 2 109 3 8 12 145 Cervidae l l l 3 Capreolus capreolus 2 3 l 2 l l lO Bison priscus l 2 3 Bos!Bison l 3 3 2 l ll Rupicapra rupicapra 5 l 5 ll Capra ibex 3 3 Total 19 lO 3 ll4 15 13 22 196 Table 2. Number of identified specimens (NISP) for ungulate taxa from the Pleistocene levels of Veternica. Tablica 2. Broj određenih uzoraka (NISP) pojedinih vrsta skupine ungulata iz pleistocenskih naslaga Veternice. est level with paleontological or archaeological remains, has been dated as Riss/Wlirm interglacial on the basis of the characteristics of the sediments and the associ­ ated remains (MALEZ,l 963, 152; MALEZ,l965, 220- 221).2 MALEZ (1965, 220) suggests that the near absence of a coarse fraction, the degraded state of the coarse fraction present, and the red color of the sedim en ts all indicate sedimentation during relatively warm climatic conditions. Furthermore, he suggested (MALEZ,l965 , 218) that calci um carbonate (CaC03) is deposited in sediments under cold-humid conditions. As such, the low percentage of CaC03 in level j (the lowest in the profile) is thought to indicate warm and dry conditions during the last (Riss/Wlirm) interglacial (MALEZ,l965, 219). Paleontologically, a warm climate was considered to be well-marked by the following species: Dicerorhinus kirchbergensis (in particular) , Dama dama, Bos primigenius, S us scrofa , Capreolus capreolus, Hystrix cristata, and Panthera pardus (MALEZ,l963, 152- 154; MALEZ,1965, 221 ). The Mousterian artifacts are purported to be "primitive" and indicative of a Riss/ Wlirm date for level j (MALEZ,1958/59, 182; MALEZ, 1963, 154). "Riss/Wlirm," as used in the works cited above, clearly refers to full interglacial conditions. In terms of the oxygen isotope record (see SHACKLETON, 1969; SHACKLETON & OPDYKE, 1973), this refers to sub-stage Se (e. 128-118,000 BP). The condition and composition of the sedim en ts from level j suggest deposition during relatively warm and humid conditions. The low frequency of the coarse fraction, the well"rounded character of the coarse fraction, and the high percentage of clays all suggest vigorous chemical weathering. While this is not the place for a detailed review of the Pleistocene sediments of Veternica, we would like to point out that the low percentage of CaC03 in level j is more likely to be indicative ofhumid instead of dry depositional conditions. A relatively low value suggests that it was leached out of the sediments due to weathering. As iron oxide (Fe203) is one of the residues of chemical weathering, one would expect higher percentages of it in levels that have witnessed relatively stronger chemical weathering. Coilveniently, level j has the highest perce n tage of iron oxide in the Veternica profile (MALEZ,1965 , 218, Table II). Furthermore, although external sources of clay cannot be ruled out (i.e. sediment brought in naturally or by human agents), it is commonly formed in place by the chemical weathering of sedi ments. It should thus indicate chemical weathering and relatively higher moisture. Chemical weathering is dependent upon moisture to operate, and as a chemical process, will increase in rate with an increased temperature. Therefore, increased chemical weathering should also indicate higher temperatures. Relative to the rest of the profile, the climatic condi­ tions during the deposition of level j were probably both warmer and wetter. This interpretation is also suggested by RUKA VINA (1983, 210), who further notes evidence of a major cooling in the upper part of the level. If one turns to the associated fauna, this revision suggests considerable attrition to the "warm climate" assemblage of level j. Fallow deer is not present, and the available rhinoceros material does not rule out the possibility that we are dealing with Dicerorhinus hemitoechus instead of Dicerorhinus kirchbergensis. Dicerorhinus hemitoechus appears to have been a more obligate grazer, and its appearance in the place of or in addition to Dicerorhinus kirchbergensis may in­ dicate the change from a temperate woody savanna to a cooler steppe environment (LOOSE,1975, 23). In any case, Dicerorhinus kirchbergensis is still present in Europe l Most of the red deer remains were not labelled by level and one cannot be sure of their exact stratigraphic provenience. However, MALEZ (J 963, J J 6) notes that the majority of red deer bones were from levels h, i, and j, and that only a few bones were _f<;>und in the upper levels (d, e, and f ). Following MALEZ's observation, these unprovenienced bones are placed in the grouped level hill]. 2 In a later work (MALEZ,J979b), he suggested that level j dates to the "upper part" of the Riss/Wiirm interglacial, although he neither gave a reason for modifiying this date nor elaborated on its chronostratigraphic implications. 8 during the early Wiirm glaciation, including remains from Hortus cave in southern France (GUERIN"l972, 207) and Ćrni kal in Slovenia (RAKOVEC,19S8, 39S- 396). Leopards have been found in a very wide range of Pleistocene contexts and are poor indicators of environmental conditions (KURTEN,l968, 88). The crested porcupine is also found in a variety of inter­ glacial and glacial contexts (KURTEN,l968, 200) and does not necessarily indicate interglacial conditions. Pig and roe deer may be more or less dependent on the presence of deciduous forest None of the taxa necessarily indicate an interglacial environment or date. Previously, the presence of bison in assemblages has been considered indicative of cold and open environ­ mental conditions. Research on fossil and recent mem­ bers of this genus, however, shows that they were adapted to a wide range of biomes and climatic conditions (BAMFORTH,l988; SALA,l986; VERESHCHAGIN & BARYSHNIKOV, 1984). The paleoecological signifi­ cance of biso n in Veternica is difficult to evaluate due to the small number and poor condition of the remains. In passing, we note that the "primitive" aspect of the artifact assemblage is very much dependent on the raw material, in this case, volcanic tuffs, basalts, quartzites, andjasper (MALEZ,l9S8/S9, 176). As MALEZ (19S8/ S9, 176) notes, these materials are all available in grave! terraces from the nearby Sava River and other locales years Oxygen BP O Isotope Stage Revised Stratigraphy (this study) Veternica Levels l 20,000 2 d e 40,000 f 3 g 60,000 h 4 i 80,000 Sa j Sb V" Sc l()() ,000 120,000 S d Se 140,000 Geologia Croatica 45 in the immediate vicinity. Furthermore, although the frequent use of quartzites in level j contrasts with raw material use at Krapina (MALEZ,l974, 16), close technological and typological similarities have been noted between the industry from level j at Veternica and the middle paleolithic assemblages from the low­ est levels at Erd in Hungary (GABORI-CSANK,1968, 2S8). The superposition of the Erd artifact levels on a well-marked paleosol indicates a date young er than the last interglacial (GABORI-CSANK,l968, lOS). The Veternica artifacts from level j need not indicate an interglacial date. Given the data presented above, we suggest that level j at Veternica does not date to the last interglacial maximum (sub-stage Se, e. 128-118,000 BP), but to a later and cooler time period with a more open envi­ ronment. In Figure S we present our suggested revi­ sion of the stratigraphy and com pare it to the existing stratigraphic interpretation by using the oxygen isotope stratigraphy as a common measure.3 The sediments, however, do indicate relatively warm and wet conditions of deposition, at least with regards to the rest of the profile. The pollen spec tra for level j (MALEZ,196S, 214) points to the presence of de­ ciduous species in the area, although such trees may have been restricted to fairly limited stands. In the absence of absolute dates, this leaves open a relatively wide time years Oxygen BP o I S so tope l 20,000 2 40,000 3 60,000 4 80,000 Sa Sb 100,000 Sc 120,000 S d 5e 140,000 tage Original Stratigraphy (Malez,1963) Veternica Levels d post glacial e f Wiirmill fl Wiinnll/111 g Wiirmll h Wiinn l/ll i Wiirml ~. j Riss/Wiirm ' Flg. S. Proposed revision of the Veternica chronstratigraphy and the original stratigraphic interpretation (MALEZ,1963, 153, Figure 12), both compared to the oxygen isotope record (after GAMBLE,l986, 76, Figure 3.3). Slika S. Prikazi revidiranog i izvornog (MALEZ,1963, 153, slika 12) kronostratigrafskog profila Veternice uspoređeni s shemom izotopa kisika (usklađeno prema GAMBLE,l986, 76, slika 3.3) 3 While MALEZ never correlated the Veternica profile with oxygen isotope stages, we note again that his emphasis on the "warm" and "interglacial" character of the fauna and sediments from level j suggests placement of this level in the warmest part of the Riss/Wiirm interglacial, sub-stage Se. ;.. Miracle & Brajković : Revision of the Un gulate ... span in which to fit level j. We favor a date during sub­ stages Sc or Sa (e. 100,000 and 80,000 BP, respectively), but will leave the question open for the time being. Our revision of the chronology of the Veternica sequence is in close agreement with a little-known in­ terpretation given by S. BRODAR (1966) in a book review of MALEZ's monographs. On primarily sedimentological and geomorphological grounds, BRODAR (1966, S19) suggested that level j dates to either a final phase of the Riss/Wlirm interglacial, or to an early Wlirm in­ terstadial. In passing, he also noted that the "clacton" technique used in making the stone tools is found in many younger sites, and that the faunal remains, as published, do not indicate an interglacial date (BRODAR 1966, S19-S20). It is satisfying to find that our data provide such clear support for an interpretation now over 2S years old. Support for the stratigraphic revision proposed above is provided by PAUNOVIĆ (1991) in a study of changes in cave bear teeth from Veternica. She observed that primitive tooth morphotypes characteristic ofRiss glacial and Riss/Wlirm interglacial deposits are not represented in the Veternica assemblages. The Veternica cave bear teeth, in both their size and frequency of morphotypes, are most similar to more evolved Wlirm populations of cave bears. Therefore, working from several in­ dependent lines of evidence, we can conclude that the Veternica sequence does not extend into the Riss/Wiirm interglacial. S. CONCLUSION The net result of Jhe paleontological revision is to squeeze the Veternica pro file into a slightly tighter time frame. The earliest level with archaeological and pa­ leontological remains, level j, could have been de­ posited during a warmer phase towards the beginning of the Wiirm glacial. Level j is followed by two lev­ els, i and h, that also contained archaeological and paleontological materials. In terms of the associated faunas and material culture, these levels are very similar to j. By level h, the crested porcupine has gone and the elk appears. Sedimentologically, these levels are very similar to each other, with the exception that the coarse fraction from h is more rounded and corroded (MALEZ,1965, 209). These levels contain relatively less clay and iron oxide, and relatively more CaC03 and rubble than level j, suggesting cooler and drier depositional conditions, perhaps during an early Wiirm stadial. A single carbon14 date from level i, of> 43,200 BP (charcoal, GrN-4984, MALEZ,l979a) more or less confirms what we already know; levels i and j were deposited before e. 45·,000 BP. This revision also has implications for the recon­ struction of paleoenvironments in the area during the deposition of levels j to h. The un gulate assemblages from layers j - h are very similar in composition. Although the indeterminate rhino and pig are restricted to level j, and elk and chamois are found only in level 9 h, it is difficult to attach any importance to these changes due to the extremely small sample sizes for these taxa. In general, the faunal assemblages from levels j to h suggest a relatively temperate forest-steppe, perhaps with patches of deciduous woodland. Together with the sites of Krapina, Vindija, and Velika pećina, Veternica provides a unique chance to moni­ tor the changing circumstances of hominid adaptation and evolution in southeastern Europe (e.g. SMITH,1982). A first step to such a general study is an adequate identification, description and interpretation of the excavated data. We hope that the present study is successful on all three counts. Our revision facilitates chronostratigraphic comparisons with other sites and adds to reconstructions of paleoenvironmental changes in the region. Although we have not addressed the role of hominids in forming these assemblages, we have established a firmer contextual foundation from which to consider such questions in the future. 6. ACKNOWLEDGMENTS We would like to thank Maja PAUNOVIĆ for access to the Veternica collections and for her support and good humor, and Jakov RADOVČIĆ for constructive com­ ments. Responsibility for the final product, however, remains our own. We also thank Slobodan OLIĆ for photographing the specimens. MIRACLE's visit was made possible by a Fulbright-Hay s Fellowship, IREX, a NSF Graduate Fellowship, and a Rackham Disserta­ tion Grant from the University of Michigan. 7. REFERENCES BAMFORTH, D. (1988): Ecology and Human Organiza­ tion on the Great Plains. - Plenum Press, New Y orle. BRODAR, S. (1966): Knjižna poročila: MALEZ, M. (1963):K vartarna fauna pećine Veternice u Medvednici. MALEZ, M (1965): Pećina Veternica u Medvednici. - Arheološki ves tnik XVII, 517- 522, Ljubljana. GABORI-CSANK, V. (1968): La Station du Paleo­ lithique Moyen d'Erd-Hongrie. - Akademiai Kaido, Budapest. GAMBLE, C. (1986): The Palaeolithic Settlement of Europe. - Cambridge University Press, Cambridge. GIFFORD-GONZALEZ, D. (1991): Examining and refin­ ing thequOOratic crown reight methxl of age estimation. -In: STINER, M.C. (Ed): Human Predators andPrey Mortality.- Westview Press, 41-78, Boulder. GUERIN, C. (1972): Le rhinoceros pleistocene de la grottede l'Hortus (Valflaunes, Herault).- In: DE LUMLEY, H. (Ed.): La Grotte Mousrerienne de l'Hortus.- E tudes Quaternaires, l, 207, Marseille. GUERIN, C. (1980): Les Rhinoceros (Mammalia, Perissodoctyla) du Miocene Terminal au Pleistocene Supćrieur en Europe Occidentale. Comparaison avec les especes actuelles. - Documents des Laboratoires de Geologie Lyon, No. 79, Lyon. KURTEN, B. (1968): Pleistocene Mammals of Europe. - lO - Weidenfeld and Nicolson, London. LEHMANN, U. (1949): Der Ur im Diluvium Deutschlandsund seine Verbreitung. -Neues J ahrbuch fiir Mineralogie, Geologie und Palliontologie Abt. B, 90, 163-266, Stuttgart. LOOSE, H. (1975): Pleistocene Rhinocerotidae of W. Europe with reference to the recent two-horned species of Africa and S. E. Asia.- Scripta Geologica, 33,1-59, Leiden. MALEZ, M. (1958/59): Das Palliolithikum der Veternica H~hle und der Blirenkult - Quartlir l O/ll, 171-188, Bonn. MALEZ, M. (1963): Kvartarna fauna pećine Veternice u Medvednici. - Palaeontologia Jugosla vica, No. 5, Zagreb. MALEZ, M. (1965): Pećina Veternica u Medvednici. l. Opći speleološki pregled. n. Stratigrafija kvartarnih taložina - Acta geologica, V, 175-237, Zagreb. MALEZ, M. (1972): Kvartargeološka i speleološka istraživanja u 1971 godini.- Ljetopis Jugoslavenske akademije znanosti i umjetnosti, 76, 215- 227, Zagreb. MALEZ, M. (1974): Noviji rezultati istraživanja paleolitika u Velikoj pećini, Veternici i Sandalji. -Arheološki radovi i rasprave, Jugoslavenske akademije znanosti i umjetnosti, 7, 7-44, Zagreb. MALEZ, M. (1979a): Prirodni okviri. -In: BASLER, D. (Ed.): Praistorija jugoslavenskih zemalja I, Paleolitsko i mezolitsko doba.- Svjetlost, 197- 219, Sarajevo. MALEZ, M. (1979b): Nalazišta paleolitskog i mezolitskog doba u Hrvatskoj. -In: BASLER, D. (Ed.): Praistorija jugoslavenskih zemalja I, Paleolitsko i mezolitsko doba. - Svjetlost, 227-276, Sarajevo. MIRACLE, P. (1991): Carnivore Dens or Carnivore Hun ts? A review of Upper Pleistocene mammalian assemblages in Croatia and Slovenia. - RAD Hrvatske akademije znanosti i umjetnosti, 458, 191-217, Zagreb. Geologia Croatica 45 PAUNOVIĆ, M. (1991): Morphometrische und Mor­ phogenetische Untersuchungen der Zlihne des Ursus spelaeus Rosenmiiller & Heinroth aus der Hohle Veternica (NW Kroatien). -RAD Hrvatske akademije znanosti i umjetnosti, 458, 1-19, Zagreb. RAKOVEC, I. (1958): Pleistocenski sesalci iz jame pri Črnem Kalu. - Razprave IV Razreda, Slovenska akademija znanosti i umjetnosti, 4, 365-433, Ljubljana. RUKAVINA, D. (1983): O stratigrafiji gornjeg pleistocena s osvrtom na topla razdoblja i njihov odraz u naslagama na području Jugoslavije. - RAD Jugoslavenske akademije znanosti i umjetnosti, 404, 199-221, Zagreb. SALA, B. (1986): Bison schoetensacki FREUD. from lsemia la Pineta (early Mid-Pleistocene- Italy) and revision of the european species of bison. - Palaeontographia Italica, 74, 113-170, Pisa. SHACKLETON, N. J. (1969): The last interglacial in the marine and terrestrial record. - Pr9ceedings of the Royal Society of London, B, 174, 135- 154, London. SHACKLETON, N. J. & OPDYKE, N. D. (1973): Oxygen isotope and palaeomagnetic stratigraphy of equatorial Pacific core, V28-238.- Quaternary Research, 3, 39-55, New York. SMITH, F. H. (1982): Upper Pleistocene hominid evolution in South-Central Europe: a review of the evidence and analysis of trends. - Current Anthropology, 23, 667-703, Chicago. VERESHCHAGIN, N. K & BARYSHNIKOV, G. F. (1984): Quaternary mammalian extinctions in northern Eurasia.- In: MARTIN, P.S. & KLEIN, R.G. (Eds.): Quaternary Extinctions. A Prehistoric Revolution. - University of Arizona Press, pp. 483-516, Tucson. WEGRZYN, M. & SERWATKA, S. (1984): Teeth eruption in the european bison.-Acta Therio­ logica, 29(9), 111-121, Bialowieža. PLATE- TABLA I lA Bos!Bison, Upper Pleistocene, Veternica level j (VTR. 3), navicular-cuboid, right, scale in cm; gornji pleistocen, Veternica sloj j (VTR. 3), naviculare-cuboideum, desni, mjerilo u cm. l B Megaceros giganteus (BLUMENBACH), Upper Pleistocene, Sandalja n level H (Sa 1772), navicular-cuboid, left; gornji pleistocen, Sandalja n sloj H (Sa. 1772), naviculare-cuboideum, lijevi. 2A Bos!Bison, Upper Pleistocene, Veternica level i (VTR. 2), phalanx UI, .scale in· cm; gornji pleistocen, Veternica sloj i (VTR. 2), phalanx III, mjerilo u cm. 2B Megaceros giganteus (BLUMENBACH), Upper Pleistocene, Sandalja II level i (Sa. 472), phalanx III; gornji pleistocen, Sandalj a II sloj i (Sa. 4 72), phalanx III. - Miracle & Brajković : Revision of the Ungulate ... PLATE -TABLA I ll 12 Geologia Croatica 45 Revizija faune ungulata i stratigarafije gornjeg pleistocena spilje Veternice (Zagreb, Hrvatska) P. Miracle i D. Brajković Paleontološka revizija fosilne faune ungulata spilje Veternice dio je opširne studije koju je započeo Pres ton MIRACLE s ciljem proučavanja tafonomija spilja Hrvatske, uz razmatranja načina i uvjeta opstanka hominidnih zajednica tijekom pleistocena (MIRACLE, 1991 ). Preliminarnim pregledom fosilne građe Veternice primjećena je određena nepodudarnost s do sada objavljenim rezultatima paleontološke obrade Veternice (MALEZ,1963), što je potaknulo potrebu za detaljnom revizijom. Rezultati revizije (tablice l i 2) značajni su za paleoekološku i stratigrafsku interpretaciju Veternice. Taksonomskom obradom utvrđena je odsutnost vrsta Dama dama i Megaceros giganteus, ali i prisustvo vrste Bison priscus. Također je izmjenjeno kvalitativno i kvantitativno sudjelovanje vrsta A/ces a/ces i Capreolus capreolus. Rupicapra rupicapra je osobito zastupljena u slojevima d i h. Capra ibex utvrđena je jedino u sloju d, dok je paleofauni sloja e pri dodana i vrsta S us sc roja. Ostaci vrste Cervus e/aphus ukazuju na pripadnost krupnim jedinkama, i ta je vrsta u cjelokupnoj fosilno j građi najviše zastupljena. Od skupine ungulata jelen dominira u gotovo svim slojevima. Do sada je sloj j, najstariji sloj koji sadrži paleonto­ loške i arheološke ostatke, po svojim sedimentološkim i faun ističkim karakteristikama bio datiran u interglacijal Riss/Wlirm (MALEZ, 1963, 152; MALEZ, 1965, 220- 221). MALEZ (1965, 220-221) je tvrdio: "toplodobni karakter sloja j vrlo dobro markiraju neki predstavnici faune, i to u prvom redu nosorog vrste Dicerorhinus kirchbergensis, zatim jelen lopatar (Dama cf.dama), dikobraz (Hystrix cristata minor), leopard (Panthera pardus ), divlja svinja (Sus scrofa), srna (Capreo/us capreolus), divlje govedo (Bos primigenius) i drugi". Isti autor smatra da sakupljeni musterijenski artefakti iz sloja j pokazuju "primitivne" karakteristike te indiciraju datiranje sloja u Riss/Wlirm (MALEZ,1958/59, 182; MALEZ,1965, 221). Sastav sedimenata sloja j sugerira taloženje u relativno toplim i vlažnim uvjetima, a revizija faunske asocijacije upućuje na određeno "osipanje" toplodobnih elemenata. Tako je primjerice utvrđeno odsustvo vrste Dama dama, a ostatke nosoroga nije moguće točno determinirati kao vrstu Dicerorhinus kirchbergensis ili možda vrstu Dicerorhinus hemitoechus. Panthera pardus i Hystrix cris tata bili su široko rasprostranjeni tijekom pleistocena čime gube vrijednost kao indikatori klimatski prilika (KURTEN, 1968, 88, 200). Sus scrofa, i Capreolus capreolus nisu karakteristične vrste za određenu ekološku nišu. Biso n priscus nekad se smatrao predstavnikom isključivo hladnodobne faunske zajednice ali danas je to mišljenje izmjenjeno (BAMFORTH,1988; SALA, 1986; VERESHCHAGIN & BARYSHNIKOV,1984). Uslijed malobrojnih ostataka bizona u naslagama Veternice teško je govoriti o paleoekološkom značaju ove vrste. Prije ustvrđena određena "primitivnost" izrade musterijenskih artefakata, zapravo veoma ovisi o vrsti materijala koji je bio upotrebljen. Artefakti iz Veternice izrađeni su od vulkanskih tufova, bazalta, kvarcita i jaspisa koji su se mogli prikupiti u šljunčanim nanosima Save i u neposrednoj blizini spilje. No neophodno je napomenuti da navedeni paleolitski materijal ne mora indicirati starost sedimenta. Po rezultatima dobivenim u ovom radu smatramo da se sloj j u Veternici ne može datirati u posljednji interglacijalni maksimum (izotop kisika, stadij Se, cca 128-118.000 BP, SHACKLETON,1969; SHACKLETON & OPDYKE,l973), već u kasniji i hladniji period s otvoren ijim biotopom. Sastav faune ungulata sličan je u slojevima j - h. Općenito sastav fa unske zajednice u ovim slojevima upućuje na razmjerno umjerene klimatske uvjete i na šumsko-stepski okoliš s rijetkim bjelogoričnim šumarcima. I u pogledu ostalog dijela profila, sedimenti ukazuju na relativno tople i vlažne uvjete taloženja. Metodom 14C datiranje sloj i na starost PLATE - TABLA II lA Bison priscus (BOJ ANUS) Upper Pleist.ocene, Veternica level j (VIR. 183), metacarpal, right, scale in cm; gornji pleistocen, Veternica sloj j (VIR. 183), metacarpus, desni, mjerilo u cm. lB Bos primigenius (BOJANUS) Upper Pleist.ocene, Šandalja II level GH (Ša 1269, 1266), metacarpal, left; gornji pleistocen, Šandalja II sloj GH (Ša 1269, 1266) metacarpus, lijevi. 2A Bisonpriscus (BOJANUS) Upper Pleistocene, Veternica level j (VTR. 196), Mt. left, scale in cm; . gornji pleistocen, Veternica sloj j (VTR. 196), M1, lijevi, mjerilo u cm. 2B Bos primigenius (BOJAN US) Upper Pleistocene, Š an dalja II BIC (Š a. 7398),level Bid, M1, left; gornji pleistocen , Šandalja II sloj BIC (Ša. 7398), M1, lijevi. ;.. :. Miracle & Brajković : Revision of the Ungulate ... • PLATE ·TABLA II 13 1 2 14 veću od 43.200 BP (ugljen, GrN-4984, MALEZ,l979a). Period taloženja naslaga teško je odrediti bez apsolutnih datiran ja. Prednost dajemo datiranjima s izotopom kisika, stadijima 5c ili 5a (cca 100 ili 80.000 BP), koji bi mogli odgovarati sloju j, no to pitanje ostaje otvoreno. Veternica, uz ostale lokalitete, Krapinu, Vindiju i Veliku pećinu, pruža jedinstvenu mogućnost promatranja svih okolnosti evolucije i adaptacije čovjeka u prostorima jugoistočne Evrope. Prvi korak prema širim znanstvenim Geologia Croatica 45 proučavanjima je odgovarajuća odredba, opis i tumačenje sakupljenog materijala. Nadamo se daje ova rasprava ispunila sva tri uvjeta. Cilj učinjene revizije je da olakša kronostratigrafske usporedbe s drugim nalazištima te da doprinese rekonstrukcijama paleoekoloških i paleo­ zooloških uvjeta tijekom gornjeg pleistocena u ovim prostorima. Premda u oblikovanju faunskih zajednica uloga čovjeka nije određena, vjerujemo da ova rasprava pruža pouzdane osnove za otvaranje nekih novih pitanja. Manuscript received March, 31. 1992. Revised manuscript accepted May, 25.1992. -