Конференция

1999-MAY16-MAY20
2nd International Mammoth Conference

ABSTRACTS (2)

edted by Jelle W.F. REUMER & John DE VOS
DISCLAIMER

The Abstracts of the 2nd International Mammoth Conference,
published in this Volume of Conference Papers, are not published
for permanent scientific record and are therefore not to be considered
as publications in the sense of Article 8 of the International Code of
Zoological Nomenclature. They merely reflect the opinion of the
author(s) at the moment of submission, and they should not be cited
in scientific works without prior written consent of the author(s).

After abstract titles (P) means poster and (L) means lecture (oral contribution)

R.T.J. CAPPERS & S. BOTTEMA - A RECONSTRUCTION OF THE LANDSCAPE ON THE BASIS OF PLANT REMAINS FROM THE MAMMOTH SITE NEAR ORVELTE (THE NETHERLANDS)

Gusel A. DANUKALOVA, Anatoly G. YAKOVLEV, Liliana ALIMBEKOVA - BIOSTRATIGRAPHIC INVESTIGATIONS OF LOWER RIVER TERRACE DEPOSITS (SOUTHERN URALS REGION) (P)

Gusel A. DANUKALOVA, Anatoly G. YAKOVLEV, Liliana  ALIMBEKOVA - BIOSTRATIGRAPHIC
INVESTIGATIONS IN CAVES OF THE WESTERN SLOPE OF THE SOUTHERN URALS (P)

I.A. DUBROVO & G. S. RAUTIAN - COMPARATIVE ANALYSIS OF GENETIC DIVERSITY IN MAMMOTH (P)

Tamara A. DUPAL - LATE PLEISTOCENE SMALL MAMMAL FAUNAS OF SOUTH WESTERN SIBERIA (P)

Marco P. FERRETTI - FUNCTIONAL ASPECTS OF THE ENAMEL EVOLUTION IN Mammuthus (PROBOSCIDEA, ELEPHANTIDAE) (L)

D. C. FISHER, D. L. FOX, & L. D. AGENBROAD - TUSK GROV^TH RATE AND SEASON OF DEATH OF Mammuthus coiumbi FROM HOT SPRINGS, SOUTH DAKOTA, USA (L)

F. A, FLADERER - DELICIOUS MAMMOTH CALVES AND WOLVES? THE CASE OF THE 27 KyBP STADIAL KREMS-WACHTBERG SITE IN THE MIDDLE DANUBE REGION (P)

I. FORONOVA & A. ZUDIN - VARIABILITY IN MAMMOTH LINEAGE AND PECULIARITIES OF ITS EVOLUTION (P)

V.E. GARUTT & V.S. BAIGUSHEVA - THE DISCOVERY OF A SKELETON OF Mammuthus trogontherii IN TERRACE DEPOSITS IN THE DELTA OF THE DON RIVER (P)

Mietje GERMONPRE - TAPHONOMY OF MAMMOTH REMAINS FROM BELGIUM (L)

Silvia GONZALEZ, Joaquin ARROYO-CABRALES, Alan TURNER, Paul PETTIT & Graham SHERWOOD - LATE PLEISTOCENE MAMMOTHS (Mammuthus coiumbi) IN CENTRAL MEXICO: PALEOENVIRONMENT
AND AMS DATING (L)

H. Douglas HANKS -ATECHNIQUE FOR THE RECOVERY OF IN SITU MAMMOTH TUSKS (P)

G. HAYNES &J. KLIMOWICZ - COLUMBIAN MAMMOTH (Mammuthus coiumbi) AND AMERICAN MASTODONT (A/lammut americanum) BONESITES: WHAT DO THE DIFFERENCES MEAN? (L)

R.-D. KAHLKE - MAXIMUM DISTRIBUTION OF LARGER MAMMALS OF THE LATE PLEISTOCENE Mammuthus-Coelodonta FAUNAL COMPLEX IN EURASIA - AREAS AND LIMITS (L)

J. KARHU& S. VARTANYAN - PALEOCLIMATIC CHANGE AT THE PLEISTOCENE-HOLOCENE
TRANSITION, WRANGEL ISLAND, EASTERN Siberia; EVIDENCE FROM OXYGEN ISOTOPES IN
MAMMOTH TEETH (P)

Hans KRAUSE - MAMMOTH FAUNA, CLIMATE, PLANT COVER AND NUTRITION (L)

Henryk KUBIAK - ADDITIONAL TEETH IN THE MAMMOTH (P)

In spring 1991 skeleton bones of four mammoths and a woolly rhino were found in situ nearby Orvelte in the northern part of the Netherlands (52њ50.52'N - 6њ41.33'E). They were recovered from a depth of 4 rn during groundwork in connection with the construction of a gas pipeline. One of the bones was radiocarbon dated to 46,800 +/-1500-1250 yBP (GrN-18780) and coincides with the Moershoofd interstadial complex, being the warmest period during the cold Pleniglacial. This date is in accordance with two other dates obtained from gyttja and peat that was collected from almost the same level. These dates indicate that all animals must have died at about the same time. Large quantities of soil in connection with the skeletons were sampled for studying plant and animal macrofossils. Additionally, one of the profiles was sampled over a length of 1 15 cm for pollenanalysis. A total of 64 vascular plants and 20 mosses could be identified on the basis of macrofossils. Together with the
pollenanalysis, an extensive record became available for the reconstruction of the former landscape.

The mammoths and woolly rhino were found in the upper course of a former brook valley. Although seeds and fruits can be transported by water over considerable distances, we therefore can assume that for this particular spot the plant macroremains represent the more local vegetation. Many plant species are indicative of both open water and bordering marshy areas and give the impression of a diverse and colourful vegetation. The presence of Potamogeton praelongus indicates that at least at some places the water must have been several metres deep. Also most of the mosses must have grown in this aquatic to wet environment. Outside this vaJley, on the moist to dry plateau an almost treeless vegetation was present. The percentage of tree pollen is very low and some of the trees may have been present in the marshy environment, such as willow (Saftx) and alder (Ainus). The most dominating tree in the pollen diagram is birch (Betula). The presence of the characteristic female cone-scales and narrow winged seeds of dwarf birch (Betufa nano) show that the vegetation must have been of a limited height. That we are still dealing with a heterogeneous landscape on the plateau is, for example, illustrated by the presence of both calcicolous species, such as Carex car/ophyliea, Scabiosa columbona and Rhytidium rugosum, and of others which are calcifuge, such as Rumex acetosella. Low temperatures are not only illustrated by the low percentage of tree pollen, but also by the presence of glacial species. In addition to the dwarf birch, the following arctic-alpine species have been recorded: Ranunculus hyperboreus, Corex chordorrhiza and Potomogeton vaginatus. Nevertheless, 'it is striking that most plant species that once witnessed the mammoths and woolly rhinos in the Netherlands, are still part of the present vegetation.

Cappers, R.T.J., Bosch, J.H.A., Bottema, S„ Coope, G.K, Van Geel, B., Mook-Kamps, E. & Woldring, H„ 1993 - De reconstructie van het landschap - in: Van der Sanden, W.A.B. et al. (eds.) - Mens en Mammoet - Archeologische Monografieen van het Drents Museum 5: 27-41

Cappers, R.T.J. & Van Zanten, B.O., 1993 - Mossen rond Orvelte over een tijdspanne van 45.000 jaar - Buxbaumiella 30:31-36

With the help of palynological, stratigraphical, faunistic methods in 1996-1998 series of sequences of lov^er rivers terraces with paleontological and archeological remains were investigated in the Southem-Urals plateau, the Over- Urals hillocky area, the Uruzan-Ai plain, Southern Urals, and the Kama-Belaja lowland (Fore-Urals). Deposits are of Middle(?), Upper (Mikulino, Leningrad, Ostashkovo horizons) Neopleistocene and Holocene.

Middle Neopleistocene, Likhvin(?) horizon. Deposits situated in the terraces socle and represented by fluvial and lacustrine sediments with rare freshwater molluscs (0.1-2.5 m thick). This time Picea-Pinus forests with small admixure of deciduous trees and treeless areas with Chenopodiaceae, Artemisia and herbage dominated in the Kuruda river valley (Over-Urals hillocky area); Pinus-Picea forests with small number of deciduous trees and treeless areas with Artem/sfa-Chenopodiaceae-herbage associations dominated in the Great lk river valley (v. Novobelokatai, Uruzan-Ai plain); Betula-Pinus forests with broadleaved trees and treeless areas with Artemisia- Chenopodiaceae-Gramineae and poor herbage biotope dominated in the Belaja river valley (v. Kaga, Southern Urals).

Upper Neopleistocene, Mikulino honzon. Deposits represented by fluvial sediments (0.7-0.89 rn thick). Betula-Pinus forests with small number of deciduous trees and treeless areas with Chenopodiaceae, Artem<aa, Polygonaceae, Gramineae, Compositae dominated in the upstream of the Belaja river valley (v. Kaga, Southern Urals).

Upper Neopleistocene, Leningrad horizon. Deposits represented byfluvial (to 1.5 m thick), lacustrine (0.25-2.35 m thick), lacustrine-subaerial (0.4 m thick; 410701570 y. LU-4149, v. Novobelokatai), eluvium-deluvium (0.3 m thick) sediments. Remains of large and small mammals, molluscs, plants and archeological finds are recorded from these sediments. Artem^o-Chenopodiaceae with iLphedra meadows-steppe associations dominated in the Tanalyk river valley (Southem-Urals plateau). The climate was dry and relatively warm. Picea, Pinus, Tsuga, Uimus, Quercus, Ainus, Betulo, Artemisia, Chenopodiaceae, Compositae grew in the Kuruda river valley (Over-Urals hillocky area). Pinus-Picea and Pinus forests with small number of deciduous trees and treeless areas with Artemisia, Chenopodiaceae, herbage grew in the Great lk river valley (v. Novobelokatai, Uruzan-Ai plain). Forests with Pinus, Picea, Ephedra with broadleaved trees and treeless areas with Chenopodiaceae, Artemisia occurred in the middle course of the Bela)a river (Kama-Belaja lowland, Fore-Urals).

Ostashkovo horizon. Deposits represented by fluvio-deluvium, lacustrine sediments of periglacial type (0,8-13.55 m thick) and fluvial sediments with freshwater molluscs (0.2-0.73 m thick) which lay with erosion on lower layers. This time vegetation represented by herbage-ArtemfSfa-Chenopodiaceae meadows-steppe associations with small Pinus-Betula-Picea forests in river valleys (Tanalyk-river). Picea, Pinus, Betula, Artemisia, Chenopodiaceae, herbage grew in the Kuruda-river valley (Over-UraJs hillocky area). Picea, Pinus, Betulo, Ainus, Tilio, Artemisia, Chenopodiaceae, Compositae grew in the Great lk and Uruzan river valleys (Uruzan-Ai plain). Pinus, deciduous
trees, Artemisia, Chenopodiaceae, Grarninea, Compositae grew in the upstream of the Belaja river valley (v.Kaga, Southern Urals). Broadleaved trees and Pinus forests with small number of Picea and poor herbaceous cover dominated the beginning of this time in the middle course of the Belaja river (Kama-Belaja lowland, Fore-Urals).

Carst caves could accumulate plants and animals remains because of their native peculiarities. This feature is of great value for biostratigraphic investigations. In 1996-98, deposits and paleontologicaJ remains were studied with the help ofpaJynological, faunistic, stratigraphical, and radiocarbon methods in caves Verhnjya', 'Zapovednaja', Atysh I ', 'Lerneza I ', 'Lemeza 2', 'Lemeza 3', 'Lemeza 4', 'Ust-Atyshskaja' in the surroundings of natural monument Atysh Waterfall' (Lemeza-river) and in the cave 'Sikiaz-Tamak-7' (Aj-river) in South Urals.
Eluviurn-deluvium caves deposits formed during the Leningrad (287001000 LU-3715, 37250 LU-3876) and Ostashkovo (12380260 LU-3861, 227501210 LU-3714) times in Late Pleistocene and Holocene. Mammal remains of Upper Paleolitic and Holocene teriocomplexes were collected from caves deposits (Sataev et at., 1998). Treeless areas were covered by meadows with herbage and Artemisia, Chenopodiaceae, Gramineae, E.fihedra and with small 'islands' with Picea, Pinus and Betula dominated in the beginning of the Leningrad time in cool climate
('Sikiaz-Tamak-7'); later forests with Picea, Pinus, Abies and Betula, Tilia, Ainus, Quercus, Ulmus dominated. Treeless areas were covered by meadows with herbage, Grarnineae, Chenopodiaceae, Artemisia ('Zapovednaja', 'Sikiaz-Tamak-7'). Broadleaved trees testify temperate-warm climate. The vegetational composition of this geological period of time was like the modern one on this temtory. Forest-steppe associations with P/cea, Betufa, Pinus, Ainus and Chenopodiaceae, Arternisia and herbage ('Zapovednaja') dominated in the Ostashkovo time. In the mountainous part of the Ai river valley the role of Pinus and Betufa increased and the role ofPicea considerably diminished in forests; Lanx and E.phedra appeared. Treeless areas were covered by herbage, Chenopodiaceae, Artemisia and a little quantity ofGramineae. Vegetation testifies of some exsiccation and a cold snap of the climate.

Forests with Betufa, Pinus and some number ofPicea, Ainus, Abies, Lanx, Uirnus, TAia, Quercus and Safix dominated dunng the Holocene ('Lemeza 1, 2, 3, 4', 'Atysh I ', 'Ust-Atyshskaja'). The part of herbaceous plants sharply diminished, it is represented by herbage, Artemisia, Chenopodiaceae and Gramineae ('Sikiaz-Tamak-7'). Small mammal remains belong to the Holocene complex. Dunng early Hotocene time (Preboreal, Boreal) open steppe landscapes dominated ('Lemeza 3'). In middle Holocene time (Atlantic, Subboreal) forest-steppe landscapes prevailed.

The following mitochondrial gene fragments were sequenced in mammoths and African and Asian elephants: I 6S nbosomal RNA of 93 base-pairs (Hoss et ai. 1994); 12S rRNA of 961 yBP (Noro et al. 1998); and cytochrome b of 277 (Hagelberg et al. 1998), 228 (Yang et oi. 1996), 1005 (Ozawa et at. 1997), and 1 137 yBP (Noro et ol.1998). The main results of these studies were opposite: mammoth was considered to be more closely related to African elephant (Noro et al. 1998, Hagelberg et al. 1998) or to Asian elephant (Yang et al. 1996, Ozawa et al. 1997). The researchers indicated that a larger number of individuals from each genera is needed to consider the evolutionary relationships of these animals. At the same time, a study involving the data on all available individuals was not performed, although the same 228 base-pair fragment of the mrtochondrial cytochrome b gene was examined in six/VI. primigenius, seven E.maximus, six L.africana, and one mastodon Mammut omericanurn (Yang et al. 1996, Ozawa et ai. 1997, Noro et at. 1998, Hagelberg et at. 1998). We calculated the pairwise genetic distancesfor these individuals and used the method of multidimensional scaling to visualize their relationships (Fig. I ). The American mastodon is distinctly isolated from the other forms and the mean distances between it and mammoths, Asian elephants, and African elephants are 6.3, 5.6, and 5.7%, i.e., approximately equal to each other (the differences are statistically nonsignificant). This appears quite natural and corresponds to the concept that mastodons diverged from the stock including common ancestor of all elephants and mammoths. The mean distances between Mommuthus-ELlephas, Mammuthus-Loxodonta, and Loxodonto-Bephos are significantly lower, 3.2, 3.3, and 4.2, respectively. At the same time these distances differ nonsignificantly from each other, and the differences between them are lower than the mean intraspecies distances in Mammuthus, Qephos, and Loxodonta (3.1, 1.0, and 1.1%, respectively). This suggests that mammoth and two living elephants are equidistant (within the accuracy of the analysis) and probably diverged simultaneously from a common stock.

It is noteworthy that the individuals of either elephant species are relatively close to each other, whereas the distance between two mammoths groups (M I -M2-M5-M6 and M3-M4) is comparable to those between theelephant and mammoth genera. This is responsible for a high intraspecies differentiation within mammoths. The distances in each mammoth group are 2.19 and 1.27%; i.e., similar to the intraspecies distances in living elephants. A high genetic differentiation (comparable to the distances between elephant genera) was also revealed for the 93 yBP fragment of the I 6S rRNA gene examined in four mammoths (Hoss et at. 1994); the distances between mammoths and African and Asian elephants differed nonsignificantly from each other. The data on 12S rRNA gene obtained for one mammoth (Noro et al. 1998) also show nonsignificant difference between the distances Mammuthus-Elephos and Mammuthus-Loxodonta (Fig. 2). Thus, the Mammuthus-Loxodonta-Elephas trichotomy should probably be resolved in favor of genetic equidistance. Moreover, a similar distance is observed between two groups revealed in mammoths, hence, we probably deal with atetrachotomy. This raises a question for further studies whether or not these two groups are distinguishable morphologically.

It should be noted that genetic equidistance does not contradict the traditional taxonomic concept considering mammoth to be closer to ILIephas. Actually, genetic distances mainly reflect the time interval passed after divergence from the ancestral stock, whereas phylogenetic reconstructions are not based on the time of divergence. They reflect the extent of differentiation among the forms, which depends on the evolutionary rates varying within an extremely wide range. Thus, genetic and morphological data reflect different aspects of phylogeny. Genetic equidistance of the four elephantid forms is evidence fortheir simultaneous divergence from each other as a result of one radiation event (or irradiation, in the sense by Kowalevsky); relatively prirnitive characters of Loxodonta suggest that (1) this divergence occurred when the common ancestor was a-t the evolutionary stage of Loxodonta or even more primitive, (2) the lineage of Loxodonta was characterized by a lower rate of evolutionary transformations, and, hence, (3) Bephos and two forms ofMammuthus acquired their advanced characters (a greater number of the tooth crests, thinner enamel, etc.) in parallel. The latter is quite probable, as recent investigations have grounded that parallel development occurs rather frequently and concerns various characters, see Rautian ( 1988).

Figure I Mammoths (M 1 -6), African (L 1 -6) and Asian elephants (E 1 -7), and American mastodon (MAS) in the space of the first two coordinate axes ofi-nuMidimensional scaling the matrix of distances based on the 228 /BP cytochrome b gene fragment (the sequences published by Noro et al. 1998, Hagelberg et ol. 1998, Yang et al. 1996, Ozawa etol. 1997 were used)

Figure 2 Mammoth (M5), African (LI-3), and Asian elephants (El-3) in the space of the firet two coordinate axes of multidimensional scaling the matrix of distances based on the 12S rRNA gene fragment (the sequences published in Noro et of. (1998) were used).

Hagelberg, E„ Thomas, M.G., Cook., Ch.E., Sher, A.V. & Baryshnikov, G.F., 1998 - Nature 370: 333-334

Hoss, M., Paabo, M. & Vereshchagin, N.K., 1994 - Nature 370: 333

Noro, M., Masuda, R„ Dubrovo, IA, Yoshida, M.C. & Kato, M., 1998-Journal of Molecular Evolution 46: 314-326

Ozawa, T., Hayashi, S. & Mikhelson ,V.M., 1997 - Journal of Molecular Evolution 44: 406-413

Rautian, A.S., 1988 - Paleontotogy as a Source of Information on the Evolutionary Laws and Factors (in Russian) Sovremennayapaleontologiya (Modem Paleontology), Moscow, Nedra2: 76-1 18

Yang, H., Golenber^ E.M. & Shoshani, J., 1996 - Proceedings National Academy of Sciences USA 93: 1 190-1 194

More than 40 localities of small mammals are known of the Upper Paleoirthic complex in the south oFWestem Siberia. There are finds oftnsectivora, Lagomorphaand Rodentiafrom deposits of the Kuznetsk Basin, the Pri-ob Plateau and caves of North-Western Altai', The list of species small mammal is given in the Table. Dicrostonyx torquatus and Lemmus sibincus are found in the Kuznetsk Basin. The distribution is limited in the north under periglacial regions. The micromammal fauna of Kuznetsk basin includes 18 species. This fauna dates to the interval (C14 dated) of 29.000-39.000 y and indicates atundra-forest-steppe environment. The late Pleistocene fauna (R-W) of the Pri-ob steppe plateau includes 10 species. The absolute age ( C14 dated) of beds is 32.000 - 31.000 y. Paleoecological analysis of this fauna shows that among rodents are dominant Lagurus lagurus, Eolagarus luteus, Spermophilus sp. They are characteristic of the steppe. The Late Pleistocene small mammal faunas from caves in North-Western Altai include 34 species. In 200-400 m altitude (cave Okladnikova, C14 age 28.000 to 45.000 y) are dominant Cricetus cricetus, Arvicola terrestns, Myospalax myospalax, M.gregalis. In 600-700 m altitude (Denisova cave, C14 age >39.000 y) are dominant M.gregalis, Lagurus lagurus, Myospalax myospalax. In > 1000 m altitude (Kaminnaya cave, the absolute C 14 age 10.000-1 1.000 y) are dominant M.gregalis, Lagurus lagurus, Alticola streizowi, Wticolo macrotis. In the modern small mammal fauna of the North-Western Altai Lagurus lagurus and Blobius taipinus are absent. The large mammal fauna ofSW Siberia includes Mommuthus primigenius, Coelodonta antiquitatis and other animaJs.

In the present paperthe functional significance ofthe change in enamel thickness and microstructure shown by the molar ofthe European mammoth lineage is discussed. A majortrend in the evolution of Mammuthus is the progressive thinning ofthe enamel ofthe molars. Maglio (1972, 1973) suggested that this trend, paralleled in several elephant lineages, is aimed to maintain an optimal distance between adjacent enamel crests on the occlusal surface, as plates, increasing in number during evolution, become more compressed. However a basically different function is also possibe. An important role performed by the occlusal structures of mammalian teeth, particularly in herbivours, is the attainment of a rather high occlusal pressure (Rensberger 1973). The critical pressure, necessary for food breakage, is controlled by the load applied by the masticatory muscles and by the area of contact between opposite teeth. In particular occlusal stress is inversely proportional to the conctact area (Rensberger 1973, Fortelius 1985). In elephants, during mastication, the enamel crests of the lower molar meet those ofthe upper one, which moves mesially in the power stroke, determining food comminution. It can be thus hypothesised that the progressive reduction ofthe enamel thickness in Mammuthus maximized occlusal stress, as the number of enamel crests increased during evolution. In order to test this possibility an enamel surface index (ESI) has been calculated for upper M3s of M. meridionalis from Upper Valdarno (Eariy Pleistocene), M. trogontheni from Sussenborn (early Middle Pleistocene) and M.primigenius from Predmosti (Late Pleistocene). The index gives a rough estimate ofthe relative area occupied by enamel on the molar occlusal surface. The mean values obtained for each species shows that, in the transition from M. meridionalis to M. primigenius, the relative enamel occlusal surface decreased considerably as a consequence ofthe enamel thickness reduction, notwithstanding an almost twofold increase in plate frequency. This suggests that the extreme reduction ofthe enamel thickness in M.primigenius was not merely a consequence ofthe marked increment in plate frequency (in order to retain a certain separation between successive crests and/or to maintain a constant total occlusal area/ enamel area ratio), but may represent an adaptation to increase occlusal pressure during mastication. Consistently, the enamel microstructure underwent a differentiation which led to a relative thickening ofthe zone with occlusaJly onented prisms (Ferretti 1998), as a probable response to the more intense abrasion caused by the increased occlusal stress. The evolution ofthe enamel in Mammuthus can be interpreted as an optimisation ofthe occlusal structures related to a progressive specialisation toward a predominantly grazing diet.

Maglio, V.J.,1972 - Evolution of mastication in the Elephantidae - Evolution 26: 638-658

Maglio, V.J„ 1973 - Origin and evolution ofthe Elephantidae - Trans. Am. Philos. Soc., n.s. 63 (3): I -144

Ferretti, M.P., 1997 - Gli elefanti del Plio-Pteistocene dell'ltalia- Doctoral thesis Universities ofModena, Bologna, Firenze and Roma

Fortelius, M., 1985 - Ungulate cheekteeth: developmental, functional and evolutionary interrelations. Acta Zoologica Fennica, 180: 1 -76

Rensberger, J.M., 1973 - An occlusion model for mastication and dental wear in herbivorous mammals - Journal of Paleontology 47 (3): 515-528

Tusks of adult proboscideans are remarkable not only in size and shape, but also in internal structure. Layers of dentin organized as a series of nested cones comprise the principal mass ofelephantid tusks and represent material deposited in a distal-to-proximal sequence, from early in ontogeny to the time of death. Complete tusks thus record much, if not all, of an animal's lifetime. Regular alternations in the conditions ofgrovvth, at several different time scales, can be traced in cross section as a hierarchical system of incremental features. These structurally distinct laminae also vary in composition, reflecting the interaction of behavioral, physiological, and environmental factors. We have analyzed dentin from nearthe growing end of several tusks ofMammuthus columbi from the Mammoth Site of Hot Springs, South Dakota, USA. Our results, described briefly below, allow us to determine rates of addition of new dentin to the tusk and to deduce the season of year when death occurred. Such data enhance our understanding of the climate and forage conditions surrounding the site and offer new insight into circumstances associated with entrapment and death of the mammoths preserved in this dramatic assemblage.

The geological context and taphonomy of the Hot Springs Mammoth Site are discussed extensively in Agenbroad & Mead (1994). The mammoth assemblage apparently represents young-adult to adult, solitary males that fell into a warm, spring-fed pond occupying a steep-walled sinkhole, having been attracted to 'its margin by water and/or vegetation that would have been accessible there even in cold months. Our initial expectation was that we might find a concentration of winter deaths or a more diffuse distnbution of deaths throughout the year, depending on the relative luxuriance of sinkhole resources and those of the nearby area.

We sampled small blocks of material borderingthe pulp cavity, nearthe pi-oximal end of tusks. On transverse thin sections, we noted regulariy-spaced zones of higher porosity that we provisionally identified as annual incremental features. Within each annual interval were about 45-55 circaseptan (ca. seven-day) incremental features, as have been recognized in tusk dentin of other mammoths (Fisher in press, Fox submitted). Plotting the thickness of successive circaseptan increments yields a profile (not shown) of the changing rate of increase in dentin thickness (apposition), with ca. weekly resolution. We also milled series of dentin samples, following circaseptan features, to examine time-series ofcompositional variation leading up to the time of death. Oxygen isotope composition of the phosphate fraction of dentinal hydroxyapatite was measured according to the method of O'Neil et o\. ( 1994). Four of the resulting oxygen isotope profiles are shown in Figure I, where conspicuous cycles with ca. 4-6 per-mil ranges, on a 6-7 mm spatial scale, record the seasonally varying oxygen isotope composition of local precipitation and plant food-water. Isotope minima are associated with the winter-spnng boundary, and maxima, with the summer-autumn boundary. These cycles each contain one high-porosity zone (black bars in Figure I, associated with high rates of dentin apposition) on the rising limb of the isotope profile. This correspondence of profiles of rate of dentin apposition, high-porosity zones, and oxygen isotope profiles corroborates the interpretation of annual featues and confirms our ability to recognize seasons and growth patterns independently of one another.

Oxygen isotope profiles, in conjunction with the profiles for rate of dentin apposition, imply a climate colderthan at present, with a relatively short warm season and a longer cold season. Our isotope results only cover the last ca. two years of life, but some of our dentin apposition profiles include earlier years. Interestingly, some of these earlier years show relatively seasonal patterns ofdentin apposition (high rates in spring-summer, low in autumn-winter) and lower annual rates of dentin apposition, but the last years of life commonly show somewhat higher annual rates with a relatively nonseasonal growth pattern characterized by fluctuations in rate ofdentin apposition independent of season. With this transition, the isotope profiles show a decreasing seasonal range, which we interpret as indicating increased use of an isotopicatly invariant source of drinking water, such as the spring-fed pond in the sinkhole, or similar effluents. The low seasonality of growth rate suggests that such individuals had developed a level of dependency on the sinkhole and its resources, remaining near it as a dependable source of food and water, and insulating themselves somewhat from seasonal shortages. While broadly compatible with earlier interpretations, the unexpected component of this scenario is the length of time over which association with the sinkhole seems to have persisted. Overall, the growth rates documented are higher than for some Full Glacial mammoths, but lower than for latest Pleistocene mammoths (Fisher in press).

With respect to season of death, Figure I shows two individuals that died just afterthe beginning of autumn (HS 00213.MSL 1 1 68) and two that died just after the beginning of spnng(MSL 1 169, HS 00281). Preliminary results from other individuals seem to accentuate these two modes, contrary to expectation. We cannot yet rule out a broad cold-season pattern of mortality, but seasonal bimodality may emerge as the pattern at this site. A possible explanation for absence of mid-winter deaths could be that the margins of the sinkhole were sufficiently near freezing to reduce the risk of slipping in (or enhance the likelihood of escape). It is also possible that the pond-margin vegetation that may have attracted mammoths to the sinkhole in spring and autumn itself died back enough in winter to reduce the likelihood ofentrapment. These results offer new constraints on interpretations of site formation processes and enrich our understanding of the conditions of mammoth growth and survival during the time prior to the Last Glacial Maximum.

Figure I Oxygen isotope profiles for final years before death ( ) ofM columbi at the Hot Springs Mammoth Site, Isotope values expressed relative to SMOW (Standard Mean Ocean Water). Analytical error < 0.2%o. Bars along x-axis mark high-porosity zones in dentin.

Agenbroad, LD. & Mead, J.I., 1994 - The Hot Spring Mammoth Site: A Decade of Field and Laboratory Research in Paleontology, Geology, and Paleoecology - The Mammoth Site of Hot Springs, SD, Hot Springs

Fisher, D.C„ in press - Season of death, grov^th rates, and life history of North American mammoths - in: West, D.L, Monotet-White, A. & Hofman, J. (eds.) - Mammoth Site Studies - University of Kansas Publications in Anthropology

Fox, D.L, submitted - Growth increments in Gomphothenum and implications for late Miocene climate change in North America - Palaeogeography, Palaeoclimatology, Palaeoecology

O'Neil, J.R., R.oe, L.J., Reinhard, E. & Blake, R.E., 1994 - A rapid and precise method of oxygen isotope analysis of biogenic phosphate - Israel Journal of Earth Science 43: 203-212

The Krems-Wachtberg srte is situated on a SE exposed slope, min, 80 m above the middle Wurmian Danube valley floor, at an altitude of 260 m. During a short campaign in 1930 less than 15 square metres of the site have been excavated. The inventory consists of 2,300 lithic artefacts, the oldest two burnt clay zoomorphous figurines from Austna, a few pieces of charcoal, which could be affiliated to Pinus with stunted growth, and c. 300 bones and bone fragments. A radiocarbon date from charcoal is 27,400 +/- 300 yBP (GrN-301 I ), The dating result, the presence of clay figurines, and the stone tool analysis show strong resemblance to the 'Pavlovian' cultural entity.

The prey species spectrum contains wolf (Cans lupus-, minimum 6 individuals), arctic fox (Alopex fagopus; 5), wolverine (Gulo gulo-, 3), woolly mammoth (Mammuthus primigenius; 9), reindeer {Rangifer tarandus-, I ), red deer (Cervus elophus-, I ), ibex (Copra ibex; I ), musk ox (Ovibos moschatus-, I ), and arctic hare (Lepus timidus-, I ). Body part abundance distributions show an overrepresentation of carnivore mandibles, with remains from at least 14 individuals. Postcranial parts form less than 10 % of the total material. Bone surfaces are highly corroded and have been destroyed by root etching and consequently only the deeper man-made cut marks are preserved. Cut marks on the occipital condyle of a wolverine skull and on the second cervical vertebra of a wolf indicate removal of the heads from the bodies. Furthermore, long bone fragments show impact marks and hinged breaks. A first analysis of the wolf remains shows similiar phenotypical, as well as cultural patterns, as the wolves from Predmosti", in Moravia. The butchering pattern shown by the carnivore remains from Krems-Wachtberg are atypical for carcass processing for mere fur acquisition and indicates an additional utilisation of the corpses for food.

Medium-sized ungulates, such as reindeer and ibex, which are under the most important prey species for the regional middle Upper Palaeolithic groups, are represented by only a piece of antler as well as a few limb fragments. The Krems-Wachtberg site is one of the few places in the middle Danube Region with finds of musk ox and its presence clarifies the stadial dry-cold climatical conditions at the time of the local camp, at c. 27,000 yBP. Mammoths dominate the animal spectrum. This species very probably played the most important role in the food procurement of the regional hunter-gatherers. The remains of at least four calves of suckling age, as well as three older calves or subadults gives strong evidence of selective carcass exploitation from a mammoth female group. At least one adult mammoth bull is represented by a nearly complete tusk Within the body parts of the juveniles, heads including isolated milk teeth are overrepresented. The osteological and the modification patterns of the entire proboscidean finds give evidence of the use of head and foot parts, as well as long bone and rib internals. Compact bone fragments were used for works and tools. From the osteological point of view, it is not possible to determine if the mammoths were killed by Paleolithic hunters or if the body parts were scavenged from a natural death site. The natural co-occurrence model of the natural mammoth-carcass accumulation and the abundance of carnivores at the c. 27,000 yBP Predmosti' site, as suggested by the German paleontologist W. Soergel, may have a similiar counterpart in the Krems-Wachtberg site. On the other hand, the hunting hypothesis may find support (1) in the dominant presence of remains of mammoth calf corpses, which probably would be under the first within a natural death site to fall a prey to any carnivore, and (2) in the multiformity and the steep relief of the regional landscape, which may have provided ambush places.

The Krerns-Wachtberg site exhibits a specialized function within a hypothetical larger camp area. Analogies exist at known Late Palaeolithic sites in Central- and Eastern Europe. Site occupation, interpreted from the mammoth calf ages, may have been during the winter months. The exploitation of animal furs and bone marrow are procedures that are of economical importance during their physiological prime state between autumn and mid-winter.

Absolon, K. & Klfma, B., 1977 - PredmostF. Ein Mammutjagei-platz in Mahren - Fontes Archaeologiae Moraviae 8: I 216,210 pis.

Eirwogerer, Th. (in press) - Die jungpalaolithische Station aufdem Wachtberg in Krems, NO - Mitteilungen der Prahistonschen Kommission der Osterreichischen Akademie der Wissenschaften, Wien

Soergel, W., 1922 - Die Jagd derVorzert - Jena

Soffer, 0., 1993 - Upper Paleolithic adaptations in Central and Eastern Europe and man-mammoth interactions - in: Soffer, 0., & Praslov, N. D, (eds) - From Kostenki to Clovis: Upper Paleolithic - Paleo-indian Adaptations - pp.31 -49. New York

Crucial new information on the history ofthe mammoth lineage was obtained. Based on abundant paleontological material, the authors created a pioneering multidimensional model ofthe lineage structure (see Figure), It illustrates discreteness ofthe evolutionary processes and considerable variability, implying a powerful genetic potential ofthe group under study. It provided flexibility and caused the mammoths to predominate among the Quaternary faunas of Northern Eurasia (Foronova & Zudin 1986, 1995, 1999). These very important results were realised while studying the main dental morphometrical features with the help ofthe authors' original approach, which was based on the building of multidimensional diagrams in coordinates of plate frequency and enamel thickness of last molars (M3). About 2,000 M3's belonging to representatives ofaJI known taxa in the lineage from numerous localities of Eurasia were used. While making samples all variability except intraspecific variability was excluded. Separate diagrams were built for Europe, Western and Eastern Siberia. They appeared to be extremely informative and clearly demonstrated the lineage structure to be far more complex than atraditional gradualistic sequence still in use.

Generally, the variability (from archaic forms to the latest mammoth) reflects canalising selection in the lineage due to global natural changes ofthe Quaternary. However, the most important and innovative result is that the selection of combinations ofthe features under study (i.e. optimality levels of dental specialisation for different stages ofthe lineage) was observed to be discrete. The structure consists of subordinated 'adaptive' peaks (zones of high distribution density) and depressions, and resembles 'Wright's symbolic picture' (Dobzhansky 1951 ). The peaks group into ensembles, so a hierarchy of boundary depressions can be seen. One ofthe large ensembles corresponds to the final stage of lineage development within the genus Mommuthus. In addition to adaptive peaks ofthe axial zone, series of peaks in 'thick-enamel' and 'thin-enamel' areas of distribution were found forthe first time. They are oppositely oriented and clinally linked with the axial zone peaks. Adaptive peaks are regarded as stages ofphenotype stabilisation (or as phenotypes), their definite positions in the lineage allowing us to correlate
them with elementary evolving structures (populations). Non-consistency of intrapopulational variability was observed both in heterochronous and isochronous populations (samples from some archaeological sites, as well as samples of contemporary African and Asian elephants). Predominance of one (sometimes two) phenotype(s) in these populations permits molars to be used for evolutionary studies, but also shows the possibility of making mistakes while working with small samples or single specimens. The fact that the structure is multi-coned implies populational variability in the lineage (polymorphism) and polytypism of taxa. Parameters of typical specimens of some taxa correspond to optima of adaptive peaks. Regional graphs are indicative of transcontinental penetration ofthe maJority of phenotypes and chiefly autochtonous speciation. Slight differences between the parameters of analogous phenotypes from various regions imply geographical clinal variability as a result of different responses of egional environments to global climatic changes. Thick-enamel and thin-enamel adaptive peaks are treated as ecological adaptations (forms). Many facts - morphofunctional differences, accompanying fauna and flora, different stations, physical dates, and a certain agreement between the sequence of these forms and the stages ofthe iriarine oxygene-isotope record - provide grounds to link thick-enamel and thin-enamel adaptations with, respectively, interglacial and periglacial environments.

Apparently, thin-enamel forms, which can be traced all the way through the lineage, played a special role. It was these forms that could pioneer new adaptive zones, since chiefly the thin-enamel phenotype was selected during evolution. The diagram shows several moments of abrupt decrease of enamel thickness before depressions - different taxonomical boundaries within the lineage (i.e., boundaries between the genera Archidiskodon and Mamrnuthus [points 13, 14, 12] and between 'early' and late' M. primigenius [points 6, 5, 4]). These boundaries are marked by their outstanding paJeogeographica) events, probably causing two large waves of station expansions: first, to the middle latitudes of Eurasia, and then to the extremely high ones.

Figure I Variability of the elephants of the Archidskodon-A^ammuthus lineage in Europe (according to M3 parameters). Coordinate axis: E - enamel thickness; PF - plate frequency on a 100 mm stretch; PL - length of one plate. Continuous isolines of distribution density are drawn through 0.5 Uniform Density Units, punctuated isolines are drawn through 0.25 Units, outer isoline corresponds to 0.25. Points are the coordinates of typical specimens oftaxa distinguished in the lineage and some peculiar forms (parameters were taten from the sources provided in Foronova & Zudin ( 1986, 1999). I. Mommuthus primigenius sifancus; 2. M. primigenius primigenius, neotype; 3. M. primigeniusjotzkovi, holotype; 4. M. primigenius fraasi, holotype; 5. M. primigenius, early form, average parameter values; 6. M. primigenius (Chokurcha site); 7. M.
primigenius, lectotype; 8. M. intemedius, holotype; 9. M. trogontheri) chosoncus, holotype; 10. M trogontherii chosaricus, holotype (authors' measurements); I I . M. trogontherii O-ogontfierii, lectotype; 12. A. trogontherii (Azov museum; authors' measurements) = A. wQsti-, 1 3. A meridionalis crornerensis, holotype; 14. A. m. voigtstedtensis, holotype; 15. A. m. tamanensis, holotype; 1 6. A. mendionalis', 17. A. rn. nneridionalis, holotype; 1 8. A. gromovi', 19. A. m. toribonensis.

Foronova, I.V. & Zudin, A.N., 1986-A new approach to the study of North Eurasian fossil elephants of the Archidiskodon-Mammuthus lineage - in: Arkhipov, S.A. (ed.) - Biostratigraphy and Paleoclimates of Pleistocene in Siberia - pp. 6-31. Novosibirsk (in Russian)

Foronova, I.V., & Zudin, A.N., 1995 - The structure of the lineage Archidiskodon-Mammuthus - in: Abstracts of First International Mammoth Symposium, 1 6-22 October 1995, St. Petersburg, Russia. Cytology 37: 648-649

Foronova, I.V. & Zudin, A.N., 1999 - The structure of the lineage Archidiskodon-Mammuthus in Eurasia and peculiarities of its evolution - in: Haynes, G., Klimovicz, J. & R.eumer, J.W.F. (eds.) - Mammoths and the Mammoth Fauna; Studies of an Extinct Ecosystem - Deinsea 6: 103-1 18

In December 1964 a skeleton of Mammuthus trogontheni (Pohlig, 1885) was discovered on the eastern outskirts of Azov city, Rostov Region, in Kagalnik sand-pit situated in the delta of the Don River on the Semibaiki terrace in a layer of bog clays. The excavations were carried out by the Museum of Regional Studies, Azov, under the guidance of V.S. Baigusheva. Recovery of the skeleton from the functioning sand-pit was performed under difficult conditions of severe winter. As a result, a large portion of the bones that were complete when extracted from the soil were damaged underthe action of warm temperature indoors. The only exception was the skull recovered in a monolith with embedding deposits. In 1982 the skeleton of the mammoth was mounted in the Museum of Regional Studies of Azov. The work was conducted underthe guidance of V.E. Garutt with the active assistance of the director of the museum A.A. Gorbenko and his colleagues. The skull was preserved almost completely. Its apex and forehead surface were slightly injured. The upper edge of the nasal opening and a part of the intermaxillary bones are destroyed. Both tusks were destroyed. Only small fragments in alveoli were preserved. Posterior molars of the excellent state of preservation include strongly worn M2 and very slightly worn M3. Bones that were preserved complete comprise 4 cervical vertebrae, right rib, 5 long bones of fore and hind limbs, 22 bones of the manus and pes. Fragments of pelvis have been discovered. The damaged bones were restored and casts were substituted forthe missing bones. The height ofthe mounted skeleton at the highest point of the vertebral column is 4.5 m. This is apparently one ofthe largest known skeletons of extinct elephants displayed in museums ofthe world. Sizes ofthe skeleton and sculpture of bones indicate that the remains belong to a male. In autumn 1998 the left tusk of a large proboscidean was discovered in the Kagalnik sand-pit immediately close to the location ofthe discovery ofthe above-mentioned skeleton. To define the identity ofthe remains the following parts ofthe skull were cleaned from clay: the nght alveole of the tusk, subocular arch and maxillary bone. It is not inconceivable that the remaining parts ofthe skeleton of Mammuthus trogontherii occur in the same location. Excavations were stopped because ofthe rains. The site was temporarily closed forthe purpose of conservation. The excavations could be continued provided that the Museum in Azov obtains financial support for this work.

In Belgium, more specifically in the Flemish Valley, large quantities of mammalian remains are found in several localities. The Flemish Valley is a paleovaltey dating from the Middle Pleistocene; it was formed by phases of fluvial and estuarine erosional activities, which alternated wrth periods of sediment accumulation. The fossils accumulated mainly through gradual, long-term processes as indicated by the scattered and dispersed spatial distribution ofthe bones in the fluvial deposits, the abundance of scavenged and subaerially weathered bones and the low numbers of carnivores. The fossil bones are from mammals that died within the confines ofthe river valleys, due to predation, disease, accident or old age. Probably most deaths occurred during winter. Food and water were limited and the weakened herbivores succumbed to the harsh conditions or fell prey to carnivores. Remains of large grazing mammals (woolly mammoth, woolly rhinoceros, steppe bison, horse, etc.) dominate the fauna) assemblages. Woolly mammoth is often the best represented species.

The richest sites ofthe Flemish Valley are Zernst IIB, dating from the very beginning ofthe Weichselian and Hofstade I, dating from the Middle Weichselian. At Zernst IIB, mammoth is represented with 747 numbers of identified specimens (NISP) and 26 minimum number of individuals (MNI) and at Hofstade I with 508 NISP and 30 MNI. The age profiles based on the mandibular dentition of woolly mammoth from the sites ofZemst IIB and Hofstade I differ from a classical attritionaJ profile as well as from a catastrophic profile. At Zernst IIB, remains from very young animals (< 6 years) are well represented. The second peak in the profile corresponds to an age group between 24 and 30 years. Fossils from very young animals are rare at Hofstade. The two peaks present are the age groups from 12 to 18 years and from 36 to 48 years.

Beside the large number of isolated marnmoth bones, a few records exists of more or less complete mannmoth skeletons. The most famous skeleton is the one found at Lier in 1860. Most of the bones belong to a bull. He stood about 3.6 m high and died when he was probably between 30 and 35 years old. Another example is the mammoth from Hoboken. The used molars points to an age of more than 50 years when this mammoth died. Calculations of the shoulder-height indicate that this was one of the smallest adult mammoths found in Belgium (2.7 m).

Weichselian mammal remains are not only encountered in the Flemish Valley but are also well represented in caves along the Meuse river and 'its tributanes. The faunal assemblages include remains of mammals that died in the caves. Furthermore, cave hyaenas, cave lions and wolves, as well as prehistoric men brought carcasses of their prey to the caves, which they used as den or dwelling. Mammoth bones are rarely found, although examples of worked mammoth ivory exist.

Mexico is particularly rich in mammoth finds (Proboscidea, Elephantidae) with localities spreading from the deserts of the North, to the rain forest ofChiapas in the South of the country. We present results from an ongoing research programme looking into the relationships between paleoenvironments, megafauna (especially mammoths) and possible human presence in the late Pleistocene/Early Holocene deposits around the Basin of Mexico. This area is rich in late Pleistocene sediments with the presence of an extinct fauna that includes mammoths, sabre-toothed cats, bison, camels, glyptodon and human remains. The sites are in an area that had a constantly fluctuating lake system, and at times may have been in closer proximity to water bodies. There are at least 2.0 localities around the Basin of Mexico where remains of mammoth and other Late Pleistocene mammals have been found. About half of these have been found in association with human artefacts (obsidian blades and Clovis-type points). The species of mammoth in this area was Mammuthus columbi, which inhabited open grassland habitats. We have re-examined the stratigraphy in three of the most important prehistoric sites around the Basin of Mexico: a) Tiapacoya, b) Tocuilaand c) Tepexpan. In these sites the sequences are mainly made up of volcanic ash and sediments derived from volcanic deposits, indicating the major impact that volcanic activity had in the area.

An AMS radiocarbon dating programme on 42 bone samples (Pleistocene fauna and preceramic man), considered the most important finds in Mexican prehistory, produced results only on 8 specimens (5 on mammoths), owing to poor collagen content, despite careful selection of the samples to be analysed. It would appear that the depositional environment was favourable forfossilisation of the megafauna but not for collagen preservation, perhaps because the sites were in the open ratherthan in caves or because the way that the specimens were conserved. The results are:
I ) Los Reyes -La Paz, mammoth (Mammuthus columbi) tibia.- 18 280 ± 160
2) Carretera a Puebia, mammoth (Mommuthus columbi) skull: 1 6 380 ± 120
3) La Villa, mammoth (Mammuthus columbi) mandible with possible cutmarks: 1 1,300 ± 120
4) Tocuila, mammoth (Mammuthus cofumbi) skull, I 1,100 ± 80
5) Hacienda de Homos, Coahuila, mammoth (Mammuthus imperator) tooth, 10,550 ± 1 10
6) San Josecito Cave, ground sloth, (Northrotheriops shastensis) rib: 19,740 ± 160
7) Santa Lucia, camel, (Cametops hestemus) mandible with possible cutmarks: 1 1,395 ± 70
8) Tiapacoya, human (Homo sapiens) skull: 9,730 ± 65

We will discuss the importance of the dates on Mexican prehistory in the Pleistocene/Holocene limit, together with the environmental changes recorded in the stratigraphy ofthe studied sites.

In June of 1998, the Science Museum of Minnesota recovered the partial remains of a mammoth that included the difficuh: removal of a waterlogged tusk that extended from near the surface to almost six feet under the surrounding matrix. The most difficult task was the decision to remove the tusk as one piece and safely transport it back to the Science Museum. An unorthodox method was adopted that required sawing the tusk from the skull section and stabilizing rt with a plywood template for travel from the site, while leaving the jacket open to permit drying and to allow the use of stabilizing agents to strengthen the tusk for future study. The purpose of this poster is to discuss the methods used in the recovery of one of the best tusks of Mammoth primagenus found in Minnesota to date as well as how these methods may be applied to other Pleistocene fossils.

Skeletons of Columbian mammoths and American mastodonts occur individually or er) masse at many Pleistocene sites in North America. Efforts to quantify the proportion of sites that contain single animals versus those containing several are hindered by limited reporting and publishing of discoveries. Initial impressions lead some paleontologists to suggest that mastodonts behaved differently from mammoths, perhaps tending to live in smaller groups or herds, or even to have been solitary animals preferring relatively thick woodlands and forests. Based on dental anatomy, bone geochemistry, pollen analyses, and paleoecological studies of fossil bonesites, it seems clear that mastodonts did segregate from mammoths in habitat choices and diets. But based on analogical studies of modern elephant (Loxodonta ofncono) biology, behavior, and bonesites, we suggest that there is no strong reason to believe that mastodont social behavior differed from that of mammoths in significant ways. The distribution of modem elephant bonesites reflects both habitat features (such as the distribution of water and forage) as well as features of elephant behavior. Mastodont and mammoth bonesites also probably reflect the same kinds of features. Allowing for differences in habitats between the North American Pleistocene contexts and the recent African contexts, it is possible to use observations and principles derived from the modem elephant studies to reconstruct important characteristics of extinct mammoth and mastodont life. We apply these observations to show that mastodonts lived in bond groups and mixed herds similar in size and demography to the groupings mammoths lived in, and we suggest that death processes affecting mastodonts were not at all distinct from the processes affecting mammoths.

The origin and dispersal of Middle and Late Pleistocene Mammuthus-Coelodonto faunas were related to the formation of a new Quaternary ecosystem, the so-called steppe-tundra or mammoth steppe. The development of the steppe-tundra occurred as a result of climatic cooling during the Early Middle Pleistocene and the consequent aridification of great areas of the northern and central Asian landmass. The increase in aridity enabled not only members of tundra-like northern habitats to disperse, but also the descendants of southern steppe forms that could cope with the decreasing temperatures, leading to an intermingling offaunal components that originated in northern as well as in central Asian habitats. In spite of a broad combination of climatic and landscape features of arctic tundras as well as continental Asian steppes within the palaearctic steppe-tundras, regional ecological differences existed depending on geographic position. These are documented in the composition of larger mammal associations.

The maps presented here of the maximum Late Pleistocene / Last Cold Stage (Devensian, Weichselian, Valdaj, Dali) distribution of characteristic larger mammals of the Palaearctic are projections of all spatial changes of habitats documented in the fossil record. The taxa under consideration are Conis lupus, Alopex lagopus, Ursus spelaeus (inclusive of smaller forms), Gulo guio, Crocuta crocuto speloea I C crocuta ultima, Panthera leo spelaea I P. leo ssp„ Mommuthus primigenius, Lquus (stenonid and caballoid forms), Coelodonta antiquitatis, Megaloceros (Megaloceros) giganteus I M. (Sinomegaceros) ordosianus I M. (S.) yabei, Alces alces, Rangifer tarandus, Bison pnscus div. ssp„ Spirocerus kiakhtensis, Saiga tatarico boreafis I S. totarico ssp. and Ovibos moschatus. From a compilation of the maximum distribution of typical members of the Late Pleistocene Mammuthus-Coelodonta faunal complex, a picture of the spatial extension of the cold tolerant fauna of the Last Cold Stage may be drawn. The main faunal assemblage was distributed from northern Spain, England, Central Italy and Greece to the Caucasus, as well as from northeast Kazakhstan, northern Outer Mongolia and northeast China to the Far East of the Russian Federation (Primorje), to northeast Siberia and parts of the shelf regions of the Arctic Ocean exposed by a fall in sea level.

The Late Pleistocene Mammuthus-Coelodonto faunas extended over about 190 degrees of longitude and 40 degrees of latitude in the northern hemisphere. In Asia there is relatively clear distinction of the N/S-extension of the distribution belts of the different taxa. Conversely, in central and western Europe, there was a nearly complete spatial (although not syncronious) overlap ofthe various elements ofthe Marnmuthus-Coelodonta faunas, owning to decreased continentality, to the more southerly spread ofthe BaHic ice sheet and to the barriers to migration caused by the mountain ranges ofthe Alps and Pyrenees. Oceanic climatic factors, an increase in continentality towards the east as well as the development ofthe arid zones of Kazakhstan and Central Asia had a noticable .influence on the configuration ofthe Palaearctic large mammal distribution ofthe Last Cold Stage.

Kahike, R.-D., in press - The History ofthe Origin, Evolution and Dispersal ofthe Late Pleistocene Marnmuthus- Coelodonta Fauna) Complex in Eurasia (Large Mammals) - The Mammoth Site of Hot Springs / South Dakota, Scientific Paper 4, Hot Springs (with general maps of Late Pleistocene distribution of cold adapted mammals from available data)

Markova, A. K. et oL, 1995 - Late Pleistocene Distribution and Diversity of Mammals in Northern Eurasia (PALEOFAUNA Database) - Paleontologia i Evolucion, Sabadell 28-29: 5-143 (with maps of well dated Late Pleistocene mammal localities)

The existence of a Holocene mammoth fauna on Wrangel Island was reported by Vartanyan et ai. ( 1993). Dated mammoth remains from the island extend from >38,000 to 3,700 yBP. The remains are uncovered from permafrost and generally very well preserved. They can be expected to provide excellent material for environmental studies relating to the Pleistocene-Hotocene transition. Oxygen isotope ratios of apatite in teeth and bones of animals reflect the oxygen isotope composition of the waterthey ingest, which, in turn, is dependent on the average temperature of the region. Mineralogically, tooth apatite is carbonate-hydroxyapatrte or dahlirte, which contains two chemically distinct oxygen components. Most of the oxygen is present as phosphate and a minor fraction as carbonate. In well preserved samples these two components can be expected to yield similar information. We determined the 5'њ0 values of carbonate (8'њ0^) and phosphate (5'њ0 ) in skeletal apatite from mammoth teeth and bones, using both new and existing sample material from Wrangel Island. Both the 8'"0^ and 8'њ0 values have a bimodal distribution related to the age of the samples. Holocene samples are enriched in 'њ0 by about 5 %o relative to the Pleistocene samples. The shift in the isotope values can be interpreted to record the terminal Pleistocene warming.

For comparison, we investigated the oxygen isotope systematics in ice samples collected in the summer of 1997 from ice wedges on Wrangel Island. Forthese too, the distribution of5'њ0 values is bimodal, resembling that for the enamel samples. There is a pronounced maximum at -21 %o (SNOW) and a less distinct maximum at about - 29 %o. Although dating of ice in ice wedges is more problematical, existing C-14 ages suggest that the former group represents Holocene, and the latter Pleistocene, precipitation. The comparison of isotope data from mammoth teeth and ice wedges is not straightforward, because the optimum for skeletal growth may be dunng the warmer time of the year, while ice wedges preserve winter precipitation. In spite of this complication these two isotope systems yield a roughly similar climatic signal implying a significant terminal Pleistocene warming.

The woolly mammoth has lived during the peak of the Last Glaciation in Yukon/Alaska in an ice-free refugium, close to the rim of the North American ice sheet, they say. The woolly mammoth has lived on Wrangel Island and Sevemaya Zemlya, up to 80њN, during the peak of the Last Glaciation, while northwestern Eurasia was covered by an ice sheet. In central Europe, south of this ice sheet, this hairy elephant grazed on loess-steppe and polar-desert, they say. Is that true? Is that scientific?

The subarctic azonal dry steppe (winter range of mountain sheep) contains now an average of only 3.8 % CP (= crude protein, dry weight), from October till March. The adult elephant would starve there to death wrth a full stomach within 3.5 months, due to lack of digestible crude protein. The adult elephants needs 8-13 % CP. The adult elephant is only able to maintain its body weight, when eating food, which contains at least 8 % CP (dry vA). The normal amount is 10-12 % CP (dry wt). The growing elephant needs 12-15 % CP. Elephant milk contains 9-15 % crude protein (dry wt) (Dierenfeld 1994). The African elephant is still able to live, where 255 gDM/m^ has grown per year from 300 mm of rain (Laws 1970). The elephant at Tsavo East starved to death with a full stomach, where only 200 gDM/m^ has grown from 235 mm of rain per year (Phillipson 1975). East African elephant calves, about 2 years old, starved to death when eating for six months food, which contained only 5.5 %
CP (dry wt). They starved to death, when their intake-deficit of digestible crude protein had reached 14.46 % of their body weight (Krause 1996, 1997).

The 2-year-old elephant calf, weighing 500 kg at Tsavo East, during the great drought of 1970-71, starved to death, when its deficit of digestible crude protein had reached 12.643 % of its body weight. Its deficit ofmetabolizable energy was then 121.018% of its body weight. The 3,000 kg adult elephant starved to death there with a full stomach, when its deficit of digestible crude protein had reached 4.128% of its body weight. Its deficit of rnetabolizable energy was then 42.795 % of its body weight. During the first dry month, their fodder contained 7.5 % CP, during the fifth dry months 2.6 % CP. And at the end of this fifth dry month, only 2 % CP or more. Up to 5,900 elephants and hundreds of black rhinos starved to death there with a full stomach (Krause 1996, 1997). The periglacial loess-steppe or steppe-tundra grows too little fodder. Annual aboveground plant production is too low (40-60 gDM/m^.yr). And during the long arctic winter it contains too little protein, about 3.8 % CP (dry wt). The elephant would starve to death, even if the grass up there were I to 2 meters tall. It cannot feed an elephant, not to mention whole herds of elephants. The woolly mammoth would starve to death within a few weeks or months, due to lack of digestible crude protein and metabolizable energy.

The azonal periglacial loess steppe in the southern Yukon Territory near 62њ N is supposed to be just like the loess steppe, on which the mammoth fauna has grazed (Geist 1971 ). But it is growing now only 61.7 gDM/m^.yr from 200 mm ppt/yr. That is 0.3085 gDM/mm/ppt per year. The winter sheep range of Yukon-Alaska contains from Octobertill March only about 3.8 % CP (dry wt) (Winters 1980). When the adult elephant is eating there as much fodder (dry matter), as it needs to maintain its body weight, it will starve to death within 3.5 months. Then the tusker has reached its critical DCP-intake deficit of 3.95 % of its body weight. Its deficit of metabolizable energy is then 40.9 % of its body weight. And it starves to death with a full stomach. That is, while eating winter grass containing only 3,8 % crude protein (dry v^t). (Krause 1996, 1997). The azonal dry steppe of Yakutia in NE Siberia is also supposed to be like the Mammoth Steppe. It ties there near the Cold Pole of the Northern Hemisphere. But it grows only 40 to 61 gDM/m^.yr. About 0.23987 gDM/mm ppt.yr is growing there.

During the height of the Last Glaciation,the Russian Plain, south of the Scandinavian ice sheet had 90 (60-120) mm precipitation per year (Softer 1990). At 0.23987 gDM/mm/ ppt.yr (as we find it now on the Yakutian dry steppe), only 22 gDM/m^.yr would grow then on the Russian Plain. And at 0.3085 gDM/mm.ppt.yr, (as we find it now on the periglacial loess-steppe of the southern Yukon), the 90 mm of precipitation per year on the Russian Plain during the height of the Last Glaciation would have grown only 28 gDM/m^.yr. Central Europe and NW Russia during the height of the Last Glaciation, south of the Scandinavian ice sheet: the loess steppe had then 121 gDM/m^.yr aboveground dry matter (living and dead) (Gliemeroth 1995). The Eurasian dry steppe contains an average 14 % of fresh green aboveground vegetation. The periglacial loess-steppe of central Europe and of northeastern Russia, west of the Ural Mountains, was growing then about 17 gDM/m^ per year. Not even the non-lactating reindeer/caribou could have lived there. It needs 25 gDM/rn^.yr. The lactating reindeer/caribou needs 35 gDM/m^.yr (Bliss 1981). The large bison bull, weighing 1250 kg, needs at least 189 gDM/m^.yr (Belovsky 1986). The African elephant is still able to live, where 255 gDM/m^.yr has grown from 300 mm/ppt.yr (marginal habitat). The elephant will starve to death with a full stomach, where only 200 gDM/m^.yr has grown from 235 rnm of rain.

Prom this I do conclude: The woolly mammoth was not adapted to arctic cold. It was not able to live on an arctic plant-cover, because there is too little fodder, and because it contains during the long winter months too little protein. The elephant would starve there to death with a full stomach, due to lack of protein and energy within a few months, weeks, or days. It would also thirst and freeze there to death. The woolly mammoth was adapted to rnild temperate climate, without an arctic winter, without ice and snow.

The assumed adaptation of the woolly mammoth to the arctic cold, is not science but only science fiction. It does not agree at all with the scientific facts, which one knows now. It should be replaced by a better explanation, one which fully agrees with the scientific facts (Krause 1978, 1996, 1997). You may find out more about this in theInternet: rne.mbers.xoom.conr'./hanskrause., under The Mammoth and the Plood' ( 1999), Condensed Version.

Belovsky, G. E., 1986- Optimal foraging and community structure implications for a guild of generalist grassland herbivores - Oecologia (Berlin) 70: 35-52

Bliss, L.C.etal., 1981 - The International Biological Programme 25, Tundra ecosystems: a comparative analysis -Cambridge, New York

Dierenfeld, E. S. 1994 - Nutrition and Feeding - in: Medical Management of the Elephant, Michigan

Gliemeroth, A. K., 1994 - PalaoOkologische Untersuchungen uber die letzten 20,000 Jahre in Europa (Paleoecological investigations about the last 20,000 years in Europe). Stuttgart

Krause, H., 1978 - The Mammoth, in Ice and Snow? Stuttgart

Krause H., 1996, 1997- The Mammoth and the Flood - Vol. 1 -3. Stuttgart

Laws, R. M., 1975 - Elephants and their habitats - Natural Environmental Research Council, Oxford

Phillipson, J., 1975 - Rainfall, primary production and 'carrying capacity ofTsavo National Park (East), Kenya - East African Wildlife Journal 13: 171-201

Softer, 0. & Gamble, C. (eds.), 1990 - The World at 18,000 YBP. London

Winters, J. F. Jr„ 1980 - Summer habitat and food utilization by Dall's Sheep and their relation to body and horn size - MSc Thesis, Fairbanks, Alaska, May 1980

In the dentition of extant mammals, as well as in extinct species abnormally developed teeth occur. Among fossil mammals teeth of the cave bear (Lfrsus spelaeus) have been studied predominantly. Abnormalities in the dentition of fossil elephants seem not to be rare either. The percentage of 'pathological' teeth in different collections of mammoth molars of Eurasia lies between 5 %to 30 %.

Acconding to several authors deformities in the dentition occur above all in species of mammals that are in danger to become extinct. However, abnormalities in the dentition of mammoth do not increase quantitatively during the Pleistocene; only in certain periods and in some areas a higher percentage of deformities of teeth may be observed. It seems that the last' Sibirian mammoths had virtually normal sound teeth. Within the European collections of mammoth teeth there was found, so far, one specimen of an additional tooth (M3) in the collection of the Institute ofPaleontology and Historical Geology in Munich (GFR).