The Eemian Mammal Fauna of Central Europe

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79 (2/3): 269-281 (2000)

The Eemian mammal fauna of central Europe

Th. van Kolfschoten1

1 Faculty of Archaeology, Leiden University, P.O. Box 9515, 2300 RA LEIDEN, the Netherlands; e-mail:[email protected] "T^ W

Received: 15 May 2000; accepted in revised form: 31 May 2000 _X_ ^1 ft

Abstract

The knowledge of the Eemian fauna of central Europe is based on the fossil record from a number of sites located in the east ern part of Germany. The faunas with different deer species as well as Sus scrofa, Palaeoloxodon antiquus, Stephanorhinus kirchbergensis and Glis glis indicate a forested environment alternating during the climatic optimum of the Eemian s.s. with areas with a more open environment inhabited by species such as Cricetus cricetus, Equus sp. (or Equus taubachensis), Equus hydruntinus and Stephanorhinus hemitoechus. Characteristic for the Rhine valley fauna are Hippopotamus amphibius and the water buffa lo (Bubalus murrensis); both species are absent in the eastern German faunas with an Eemian age. Taking into account the short period of time covered by the Eemian s.s., the amount of data on the Eemian mammalian fauna is remarkably large. There is, however, still an ongoing debate on whether the stratigraphical position of a number of faunas are of Eemian or 'intra-Saalian' age. Furthermore, there are faunal assemblages or stratigraphically isolated finds re ferred to the Eemian without indisputable evidence. This is particularly the case in the Rhine valley, where most of the so- called Eemian fossils come from dredged assemblages. The picture of the evolution of the Eemian fauna and its geographical variation is consequently still incomplete.

Keywords: central Europe, Eemian, mammalian faunas, Northwestern Europe, Pleistocene

Introduction the Eemian stage, the Riss-Wurm interglacial or, in the British Isles, the Ipswichian), the interval following Important in the debate on present-day climatic the penultimate (Saalian/Riss/Wolstonian) glacial ice- changes are the solid data from earlier interglacials advance and preceding the last (Weichselian/ Wurm/ that can be deduced from the fossil botanical and Devensian) glacial stage. Nowadays, it is known, ho zoological records. The last interglacial in particular is wever, that the Eemian stage - as defined at the type important since the set of data is rich compared to locality near Amersfoort (the Netherlands) - covers that of earlier interglacials. This is also the case for the only a restricted part of the interval between the two mammalian record and - although the mammalian glacial ice-advances. The Eemian in a strict sense is the fauna differs in many aspects from the present-day first warm/temperate episode after the retreat of the fauna of NW and central Europe - the fossils may Saalian continental icecap; it is broadly the continental contribute to the reconstruction of the environment equivalent of the marine isotope substage (MIS) 5e and its evolution during the earlier part of the Late (see Turner, 2000 - this issue). Several temperate in- Pleistocene. terstadials separate the Eemian s.s. from the latest Many faunas have been referred to the last intergla (Weichselian) glacial maximum and part of the fossil cial (which is known in western and central Europe as remains, previously referred to the Eemian, might date

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 269

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 from these interstadials. In the present contribution, is, however, sometimes missing. In order to avoid cir the author focuses on the Eemian s.s. and in particular cular reasoning {Hippopotamus indicates an Eemian on the mammalian record from this time. age; Eemian faunas are characterised by the presence When discussing the Eemian s.s. fossil record, it is of Hippopotamus), the author has decided to restrict important to realise that there are no mammalian re the analyses of Eemian faunas to those associations mains from the type locality of the Eemian nor from that can be correlated with the Eemian s.s. and to de Eemian deposits in the type region, the Amersfoort fine the Eemian fauna in the various regions on the Basin. Correlations between the type deposits and basis of sites with dating evidence independent of the mammalian faunas are often based on palynological mammalian fauna itself, be it lithostratigraphic or ra data, absolute dates and/or the stratigraphical posi diometric. tion of the levels that yielded the fossil remains. A Central Europe (in particular the eastern part of number of faunal remains have been referred to the Germany: Fig. 1) is of major interest for the study of Eemian because of the 'interglacial' character of the Eemian faunas. Localities in this area have yielded the fauna; solid independent evidence for an Eemian age most extensive record. A general picture of the Ee-

• Lehringen

Schflnfeld

Bikgtonna' •Taubach

• Stuttgart UnterturKh§im

0 50 100 km

1HJ Weichselian continental ice-sheet I I Saalian continental ice-sheet

Fig. 1. Location of the sites with a clear stratigraphical setting, where mammalian remains with an Eemian s.s. age have been collected.

270 Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 271

to Table 1. List of mammalian species represented in the Eemian s.s. assemblages from sites in central Europe,

https://www.cambridge.org/core to Lehringen Grobern 1 Grabschiitz Neumark- Rabutz Schonfeld Taubach Burgtonna Stuttgart-Unterturkheim Nord

REPTILIA Emys orbicularis x x x x x x . https://doi.org/10.1017/S0016774600021752 MAMMALIA Insectivora

. IPaddress: Erinaceus europaeus x x Talpa europaea x x x x x x Sorex araneus sp. ex gr. ex gr. ex gr. cf. x x

170.106.35.168 Sorex minutus Neomys fodiens x Neomys anomalus cf. Crocidura suaveolens cf. sp. sp. , on

05 Oct2021 at10:51:29 Lagomorpha Lepus timidus Lepus europaeus x cf. sp. x 0 Rodentia Glis glis x Cricetus crieetus sp. x x , subjectto theCambridgeCore termsofuse,available at Cricetus major x Castor fiber x x x x B g" Clethrionomys glareolus x x x x x 5 Arvicola terrestris sp. sp. x x x x ij Microtus arvalis x x x S Microtus agrestis x x Sj M. arvalislM. agrestis sp. sp. ex gr. x a Microtus subterraneus x x o. 01 Microtus oeconomus x

t —< c Apodemus sylvaticus sp. sp. sp. sp. x x § Apodemus flavicollis cf. 0 o^ Apodemus maastrichtiensis cf. x Carnivora Canis lupus x x cf. x x x H

X X X X X X X

a 3 R -e -« 1 a s a •& -fe .2 a •§ g 5 t, 5 s£a •1 * § 1o I J J I 1 S -a. R «5 *3 •1a} sE s"-J -j~?E ~S *• *• » *-

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 Neumark-Nord (Sachsen-Anhalt) imply that the faunal remains are of 'intra-Saalian', A significant interglacial flora and fauna has been i.e., pre-Eemian, age. collected from lacustrine deposits exposed in the open-cast lignite mine at Neumark-Nord. The lacu Schonfeld (Brandenburg) strine sediments represent a small lake accumulation At the Schonfeld locality (near Cottbus), east of the formed in the Saalian till deposits that cover the localities mentioned above, a sequence with Eemian area; they lie on top of Saalian (Drenthe) till and deposits yielded a mammalian assemblage that is cor meltwater deposits. The interglacial deposits are related palynologically with the Eemian pollen zones topped by a Holocene soil and a periglacial succes E4 - E5 (Heinrich, 1991). The lacustrine Eemian de sion that can be subdivided into upper and lower posits were formed in a deep depression in the Saal loesses, which are separated by, for instance, an al ian III till. At the southern margin of the basin, a suc ternation of humic gley deposits. In the lower part of cession covered the late Saalian until the early Weich this succession well-developed ice wedges occur. selian. This sequence yielded palaeobotanical remains There is no doubt about the age of the upper part of as well as molluscs and a large number of vertebrate the periglacial sequence: all authors refer this part of fossils. The vertebrate fossil assemblage represents the succession to the Weichselian. There is, however, fishes, amphibians, reptiles (among them Emys orbicu a debate on the age of the lower part of the sequen laris), birds and twenty species of mammals (Bohme, ce. Mania (1992) assumed that the lower part was to 1991; Fischer, 1991; Heinrich, 1991). be correlated with the Saalian (Warthe substage); the lacustrine deposits have an intra-Saalian age in The debate on the age of the interglacial deposits at Grab- this scenario. Litt (1990) argued for an Eemian age schiitz, Neumark-Nord and Rabutz based on palynological data. This discrepancy will be discussed later on. The geological setting at the Grabschutz, Rabutz and The mammalian record from Neumark-Nord is re Neumark-Nord localities is more or less identical and markable because of its unusually large number of similar to the one at Grobern, i.e. interglacial deposits more or less complete skeletons of large mammals in a basin formed within glacial deposits referred to (Mania, 1992; Pfeiffer, 1998): Palaeoloxodon antiquus the Saalian (Drenthe) ice advance. According to Litt (3), Stephanorhinus kirchbergensis (1), Dama dama (1990,1994), the interglacial deposits are, as indicated (80), Cervus elaphus (18) and Bos primigenius (1). In by palynological data, of Eemian age. Mania (1999) addition, there is a collection of smaller vertebrates excluded an Eemian age, however, and preferred to and isolated remains of, among others, Emys orbicula correlate the Neumark-Nord interglacial deposits with ris and carnivores such as Ursus arctos, Panthera leo an 'intra-Saalian' warm phase, post-dating the older and Canis lupus. Drenthe and pre-dating the younger Warthe ice advan ces. This interglacial had apparently subcontinental, Rabutz (Sachsen-Anhalt) warm climatic conditions, as shown by both the ma- One of the classical sites with an Eemian fauna is Ra lacological and the palaeobotanical record. Von Koe- butz, between Halle and Leipzig. The first publication nigswald & Heinrich (1999) also preferred to correlate reporting the discovery of mammalian fossils from the the faunas from Grabschutz and Neumark-Nord with clay deposits at Rabutz is from Von Fritsch (1880); he this 'intra-Saalian' event because of the occurrence of described the upper dentition of a rhinoceros, Ste Apodemus maastrichtiensis in both faunas; this species is phanorhinus kirchbergensis. More fossils (including absent in the faunas from Grobern, Taubach and Emys orbicularis) have since been collected during an Burgtonna. The fauna from Rabutz might, according excavation in 1914; Soergel (1920) has described the to Von Koenigswald & Heinrich (1999), also date from collected fossil material. A review of the geological the 'intra-Saalian' interglacial. setting, the fossil flora and fauna has been published, The evidence for an 'intra-Saalian' interglacial se together with a description of the Palaeolithic arte parating the Drenthe and Warthe substages is, howe facts, byToepfer (1958). ver, still very poor and contradictory (Litt & Turner, The Rabutz basin with the clay deposits yielding 1993). Undeniable interglacial deposits sandwiched the mammalian fossils is a trench in the Saalian till between the Drenthe and Warthe tills have never been plateau (Eissmann, 1990). Eissmann (1990) referred demonstrated so far and a sequence with the 'intra- the interglacial clay deposits to the Eemian. There is, Saalian' (post-dating the Drenthe substage) and Ee however, a debate on whether the clay is covered by mian interglacial deposits in superposition is as yet colluvial deposits or by a late Saalian till as stated by, also unknown up to now (see also Turner, 2000 - this among others, Erd (1990). The second option would issue).

274 Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 Travertine sites Eemian age for the main travertine section at Weimar- Ehringsdorf. Heinrich (1982) also assumed a pre- In the unglaciated area not far from the German sites Eemian, i.e. late Middle Pleistocene age, for the fauna mentioned above, a few localities are present where from Weimar-Ehringsdorf because of the evolutionary vertebrate remains with an assumed Eemian age have stage of the watervole Arvicola terrestris. It is, how been recovered from travertine sequences. The exact ever, not clear whether the Lower and Upper Traver chronostratigraphical position of the faunal record is, tine, separated by the Pariser horizon, date from a however, not always clear. A direct link with the Saal- single warm/temperate episode or clearly differ in age. ian Drenthe I and II and Warthe substages, as obser The Lower Travertine might date from the Middle ved at the sites mentioned above, is missing at these Pleistocene, whereas the Upper Travertine and the travertine sites. Other methods, for example radiome Pariser Horizon might date from the Late Pleistoce tric dating, must therefore be applied. A second pro ne. The situation at Weimar-Ehringsdorf might be blem is that these assemblages are often from old col even more complex, due to karstic fissure infillings in lections: fossils collected during the exploitation of which younger (Eemian?) sediments (including fossil the travertine many years ago. Detailed information remains) penetrated into the older travertine deposits about the exact provenance of many specimens is (Mania, 1997). Other authors have questioned this missing and one can try only to reconstruct their ori concept (Kahlke, pers. comm., 2000). ginal stratigraphical position by combining present It is obvious that the situation at Weimar-Ehrings knowledge of the exposures with information from dorf is far from clear. Therefore only the fauna from the old literature. This is the case for the German sites Taubach is included in the analyses; the faunas from inThuringia (Thuringen) near Weimar and in Baden- Weimar-Belvedere Allee and Weimar-Ehringsdorf are Wurttemberg (near Stuttgart), where travertine depo ignored. The Eemian assemblage from Burgtonna, a si sits of Eemian age were or still are exposed. te at about 50 km west ofWeimar where Late Pleistoce Well known are the travertine deposits near Wei ne (Eemian and Weichselian) faunas have been collec mar. Three geographically isolated Pleistocene traver ted, is also included. Far beyond the area, in southern tine occurrences, in which the sites Taubach, Weimar- Germany near Stuttgart, is another travertine site cal Ehringsdorf and Weimar-Belvedere Allee (=Weimar led Stuttgart-Unterturkheim, which has yielded an ex Stadt) are located, yielded mammalian fossils. The tensive assemblage of Eemian vertebrate remains. fauna from Taubach is one of the classical Eemian faunas. The mammalian fauna from Weimar-Belvede Taubach (Thuringen) re Allee, also referred to the Eemian, partly resembles During the second half of the 19th century, intensive the fauna from Taubach; the faunal assemblage from extraction of travertine took place at the village of Weimar-Belvedere Allee is a mixed assemblage, howe Taubach, about 3 km southeast ofWeimar. The majo ver, and the exact stratigraphical origin of several spe rity of the palaeontological and archaeological finds, cimens, including the remains of Mammuthus primige- including two hominid teeth, were collected between nius, Stephanorhinus hemitoechus and Rangifer tarandus, 1870 and 1900 (Kahlke, 1994).The geological setting is unknown. Furthermore, independent age indica of the Pleistocene deposits at Taubach is rather com tions for the travertine deposits are missing. The fau plex and differs over short distances (Steiner, 1976). na has therefore not been integrated into the analyses. The general, simplified Pleistocene section shows Even more problematic - and heavily debated for Saalian gravels of the Ilm river at the base, covered by many years - is the age of the sequence exposed at more fine-grained flood-deposited loams. On top of Weimar-Ehringsdorf. The geological setting of the these deposits occurs an alternation of thin plates of travertine deposits at Weimar-Ehringsdorf is roughly compact travertine and less compact sandy travertine comparable to that of the sequence at Taubach; both layers, some of which are rich in molluscs. Halfway travertine successions were deposited on gravels of up the section an up to 1.5 m thick compact traverti the Ilm river and have an assumed Saalian age. The ne bed (the so-called 'Werkstein Travertine') is pre sequence at Weimar-Ehringsdorf differs, however, in sent, with on top the lower humic travertine sands, a some aspects. The so-called 'Pariser Horizon' is mis level that yielded flint artefacts and larger mammal sing in the Taubach sequence but more important are bone fragments as well as a small number of smaller the differences in the malacological assemblages from mammals. The layer with artefacts is covered by more both sequences; the Helicigona banathica association, massive travertine layers, the upper humic travertine known from Eemian deposits at Taubach and Burg- sands and, again, an alternation of thin plates of com tonna, is missing in Weimar-Ehringsdorf. Mania pact travertine and less consolidated sandy travertine (pers. comm., 2000) therefore postulates a pre- layers rich in molluscs.

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 275

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 The stratigraphical origin of the major larger mam species occur with less than twenty specimens. mal assemblage, collected during the end of the 19th The uppermost part of the travertine sequence has century, is not clear. Steiner (1976) assumed that the yielded a small fauna with, among others, a porcu so-called 'Knochensand', which yielded the larger pine, Hystrix cf. vinogradovi (Maul, 1994).The level is mammal fossils, should be correlated with the lower correlated with a late interglacial phase. In the Early part of the sequence, the section between the flood- Weichselian sequence, two levels with mammalian re deposited loams at the base and the Werkstein Traver mains have been recorded; the lowermost is characte tine. The malacological record from the sequence ex rised by the occurrence of Spermophilus citelloides, the posed at Taubach does not show hiatuses; the record upper level by the presence oi Lagurus lagurus. is comparable to that from the Eemian travertine sec The Eemian age of the main travertine deposits at tions at Burgtonna. Burgtonna is primarily based on the interglacial char The correlation between the mammalian fauna and acter of the floral and faunal assemblages from the the Eemian interglacial was initially based on the dis deposits and the evolutionary stage of mammalian tinct interglacial character of the fauna and its litho- species (such as Arvicola terrestris). The correlation is stratigraphical position. Absolute dates confirmed the supported by radiometric dates of 104-111 + 7 ka earlier assumptions. The travertine deposits at Tau (Brunnacker et al., 1983). bach have been radiometrically dated to 116 + 19 ka (Brunnacker et al., 1983). The fauna from Taubach is Stuttgart-Unterturkheim (Baden-Wurttemberg) very divers; the fossil assemblage is, however, domina Several travertine outcrops occur in the area near ted by Ursus arctos (number of specimens, n, = 1557), Stuttgart. The outcrop at Stuttgart Bad Cannstatt, at Stephanorhinus kirchbergensis (n = 1224), Bison priscus the left bank of the Neckar valley, is known for its (n = 532), Castor fiber (n = 323), Cervus elaphus (n = Middle Pleistocene fauna. The travertine of Stuttgart- 207) (Bratlund, 1999). All other species are less fre Unterturkheim, on the right bank of the Neckar val quent The presence of Stephanorhinus hemitoechus in ley, has yielded a Late Pleistocene fauna correlated the main fauna from Taubach, listed in Von Koenigs- with the Eemian stage. The reader is referred for mo wald & Heinrich (1999), for example, is not certain; re detailed information toWenzel (1996) and others. this species is represented by an isolated molar, the The travertine sequence at Stuttgart-Unterturk provenance of which is unknown (Kahlke, 1977). heim is divided in a consolidated Lower Travertine Taphonomical processes (in particular the activity of and an unconsolidated Upper Travertine by a dark- hominids) affected the composition of the faunal as coloured, humic layer, the so-called 'Steppennager- semblage, as can be deduced from the age profiles of schicht'. The vertebrate fauna from the latter level is the dominant species and the cut-mark frequencies rich in species; the presence of, among others, Ocho- (Bratlund, 1999). tona pussila, Lagurus lagurus, Alactaga major, Mammu- thus primigenius, Coelodonta antiquitatis and Rangifer Burgtonna (Thiiringen) tarandus indicate a 'mammoth steppe' biotope for this The travertine deposits at Burgtonna overlie Saalian level. Wenzel (1996) correlated the Steppennager- fluvial terrace deposits; they are, in turn, overlain by schicht with the early Weichselian Herning stadial; he Weichselian loess. The travertine complex has a thick correlated the Upper Travertine with the Brarup in- ness of about 15m and is composed of several, rather terstadial. well stratified layers. The travertine succession is very Only the Lower Travertine is of Eemian age; the rich in molluscs: 140 species have been recorded fauna listed in Table 1 originates from the Lower Tra (Mania, 1973). The mollusc fauna from the upper vertine deposits. The Lower Travertine fauna is rich in part of the travertine succession is characterised by carnivores; dominant in the assemblage are the re the occurrence of Helicigona banatica, indicating mains assigned to Felis sp. (n = 46) (Wenzel, 1996) (= warm and humid climatic conditions. The floral re Felis sylvestris as listed by Von Koenigswald & Hein mains (also well represented in the travertine depo rich, 1999). Species that are also well represented in sits) indicate a mixed oak forest with thermophilous the fauna are Cervus elaphus (n = 18), Panthera leo Mediterranean and Atlantic species. Fishes, a herpe- spelaea (n = 13) and Vulpus vulpus (n = 8). The Lower tofauna, including Emys orbicularis (Bohme, 1989) Travertine has also yielded a floral record; it indicates and birds as well as a diverse mammalian fauna, have interglacial conditions (Schweigert, 1991). been collected from these travertine deposits. The The assumed Eemian age of the Lower Travertine mammalian fauna assemblage presented by Kahlke is supported by the results of U-series dating (Griin (1978) is dominated by Cervus elaphus (n = 108); Bi et al., 1982). The three samples from the Lower Tra son priscus is also well represented (n = 22).The other vertine gave ages of 133 +21/—17 ka, 106 +6/-6 ka

276 Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 and 105 +91-1 ka, respectively, whereas sample 4, ta assemblages also contain remains of Middle or even ken from the base of the Upper Travertine (20 cm Early Pleistocene age. The sections in the Rhine valley above the Steppennagerschicht) gave an age of 105 show huge hiatuses in several places between Late +4/-4 ka. Pleistocene sediments and those beneath. One can not, therefore, exclude that part of the fossil record The Rhine valley and other sites (including warm/temperate indicators) does not date from the Eemian interglacial. It is not accepted there Distributed over northwestern and central Europe fore to date the remains of the giant beaver, Trogonthe- there are several other localities where fossil remains rium cuvieri, the small rhinoceros, Stephanorhinus sp. with an assumed Eemian age have been collected. A (= Stephanorhinus cf. megarhinus in Von Koenigswald, number of these sites are located in the valleys of the 1988), and Alces latifrons as Late Pleistocene or even Rhine and its tributaries such as the IJssel (the Eemian, as suggested by Von Koenigswald & Menger Netherlands) where, at numerous places, sediments (1997), Pfeiffer (1999) andVon Koenigswald & Hein- have been dredged up. Large numbers of fossils, in rich (1999). The giant beaver and the small rhinoce particular mammalian remains were also dredges. ros are, when collected from a clear stratigraphical The bulk of the dredged sediments date from the La context, only known from the Middle or Early Pleis te Pleistocene and Holocene; hence, the majority of tocene. Alces latifrons is also known from several the mammalian remains are of Late Pleistocene or Middle Pleistocene localities (Bilshausen, Mosbach, Holocene age. The mammalian record from these si Voigtstedt, to name a few), but its stratigraphical ran tes is composed of a combination of remains from do ge might be larger if the assignment of the antler frag mesticated animals and extinct species such as the ments from Taubach to Alces latifrons postremus is cor woolly mammoth. These mixed assemblages are rect. This does not imply that the A. latifrons remains known from many gravel pits (Grofi-Rohrheim, Eich from the Rhine valley also date from the Late Pleisto and Leeheim) in the upper Rhine valley west of Hei cene. They have been heavily mineralised, more than delberg (Germany) and from pits along the IJssel near the remains from the same assemblage that were as Zwolle (the Netherlands). The mammalian record signed to Alces alces by Pfeiffer (1999). This supports from these sites shows the presence of species indica the assumption that the remains of Alces latifrons tive of glacial conditions (such as Mammathus primi- might also pre-date the Eemian. genius, Coelodonta antiquitatis, Rangifer tarandus), as It is obvious that it is not so easy to unravel the well as species that inhabited the area under mixed, dredged assemblages from localities in the warmer/temperate conditions (such as Palaeoloxodon Rhine valley and to divide the list of species represen antiquus, Stephanorhinus kirchbergensis, Hippopotamus ted in the assemblages into stratigraphically natural amphibius, Bubalus murrensis). The former species are units, and to filter the purely Eemian faunal elements. assigned to the Weichselian, whereas the latter group This also applies to other mixed assemblages, for in of species is dated as Eemian. stance those from the Maasvlakte near Rotterdam Although it is obvious from additional (palaeobota- and the Eastern Scheldt estuary where part of the nical and geological) research that part of the dredged mammalian fossils might have an Eemian age. sediments and hence, most probably also part of the There are also a number of localities in northwes fossil remains, date from the Eemian, it is still unclear tern and central Europe where mammalian remains which part of the assemblage should be referred to of assumed Eemian age have been collected from the Eemian. Because of the ecological demands of the clear stratigraphical contexts. One of these sites is hippopotamus, Hippopotamus amphibius, and the wa Scladina Cave near Sclayn (Belgium), where a level ter buffalo, Bubalus murrensis, which both cannot tole with, among other species, Canis lupus, Vulpes vulpes, rate periods of severe or long-lasting frost, one may Alopex lagopus, Panthera (Leo) spelaea, Crocuta crocuta assume that both species date from the Eemian (Von spelaea, Ursus arctos (the dominant species), Ursus spe- Koenigswald, 1988; Van Kolfschoten, 1995). This is, laeus, Equus sp., Dama dama and Homo neanderthalen- however, less obvious for other species of the warm/ sis has been assigned to the last interglacial. The abso temperate group. One cannot exclude that the re lute dates confirm the early Late Pleistocene age, the mains of, for example, Palaeoloxodon antiquus and dates varying from 100 ka to 127 +46/-32 ka (Boche- Stephanorhinus kirchbergensis exclusively date from the rens et al., 1999). Palynological correlations suggest, Eemian s.s. and not from the temperate Early Weich however, that the level corresponds with the Early selian interstadials. Weichselian Saint-Germain II interstadial (marine Another problem with the fossil assemblages from isotope substage 5a) (Bastin, 1992). the sites along the Rhine is the fact that many of these The localities Tonchesberg (Van Kolfschoten &

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 277

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 Roth, 1995) and Wallertheim (Conard et al., 1995) & Heinrich (1999), among others, grouped faunas has also yielded fossil remains that have been dated as from Rhenen-Leccius de Ridder and Maastricht- Eemian. It is obvious that part of the sequence dates Belvedere (the Netherlands) and Hunas, Weimar- from the Late Pleistocene, but the Eemian age of spe Erhringsdorf (Germany) and assumed a pre-Eemian cific levels is, however, mainly based on the 'best fit' 'intra-Saalian' age for them. The faunas are characte method. The mollusc associations from Wallertheim rised by the occurrence of Arvicola terrestris molars show similarities with the associations from the upper with an evolutionary stage roughly intermediate be part of the last-interglacial sequence at Burgtonna, in tween that of the Middle Pleistocene Arvicola terrestris dicating that the lowermost mammalian find-layer at cantianus from the Reinsdorf interglacial deposits at Wallertheim dates from the later part of the Eemian Schoningen for example and the modern central Eu (Mania, unpubl.).The amount of data supporting the ropean Arvicola terrestris populations. Von Koenigs Eemian s.s. age is, however, restricted and an early wald & Heinrich (1999) excluded an Eemian age for Weichselian age cannot be entirely excluded. This the faunas from Neumark-Nord and Grabschutz be leaves these faunas less applicable for the unambi cause of the occurrence of Apodemus maastrichtiensis. guous determination of the Eemian mammalian fau The first question, however, is whether all these 'in na from central and western Europe. tra-Saalian' faunas date from the same warm/tempe Another locality with a mammalian fauna dated as rate event. Eemian is Steinheim an der Murr (Von Bloos et al., The term 'intra-Saalian' is confusing. Mania 1991; Von Koenigswald & Heinrich, 1999).The smal (1997) used the term 'intra-Saalian' to indicate a ler vertebrate fauna comes from slope deposits and warm/temperate interval post-dating the Drenthe tills badger borrows. The fauna, mainly with voles but also and pre-dating the War the tills. In his concept, the - although rare - Apodemus, Cricetus and Spermophilus Saalian begins with the Saalian I (Drenthe) ice ad citelloides, does not indicate full interglacial conditions vance; he referred to the interval between the Elste- and points as a whole (including gastropods) to early rian and the Saalian I as 'Holsteinian Complex' (Ma Weichselian interstadial climatic conditions (Von nia, 1997). Other authors, however, use the term Bloos et al., 1991). An Eemian age for the fauna has, 'Saalian' to indicate the time between the Holsteinian however, been put forward because of the evolutiona s.s. and the Eemian. 'Intra-Saalian' might, in this con ry stage of the Arvicola molars, despite the fact that cept, also mean post-Holsteinian and pre-Saale I. The the remains come from deposits stratigraphically abo faunas from Rhenen-Leccius de Ridder andWagenin- ve the 'Eemian' soil. According to Heinrich (in Von gen-Fransche Kamp (Van Kolfschoten, 1990) also Bloos et al., 1991), the enamel differentiation of the have an 'intra-Saalian' age, according to this concept; molars from Steinheim is less advanced than that of the sediments in which they were found have been the molars from the Stuttgart-Unterturkheim Eemian pushed by the Saalian (Drenthe) ice sheet and hence fauna. The differences are small: statistically too small pre-date the Drenthe glaciation. These faunas must to overrule the lithostratigraphical and ecological da be older than the fauna from Neumark-Nord, for ta. Furthermore, the evolution of the watervole, Arvi example, collected from deposits on top of the Dren cola, is not so straightforward (Van Kolfschoten, the till. It is therefore assumed that the 'intra-Saalian' 1990). Because of the discrepancies in the age indica faunas are not contemporaneous and should be divi tions, it has also been decided not to include the ded into a group of pre-Drenthe I faunas (e.g., Rhe Steinheim fauna in the analyses. nen-Leccius de Ridder, Maastricht-Belvedere, Wage- ningen-Fransche Kamp, Hunas, Weimar-Erhrings- dorf) and a group of'Eemian' faunas (e.g., Neumark- Discussion Nord, Rabutz and Grabschutz). Theoretically these faunas could date from Mania's pre-Eemian, pre- Current knowledge of the Eemian mammalian fauna Warthian 'intra-Saalian' interglacial, but there is no from the western and central part of continental Eu convincing evidence for such a warm/temperate phase rope is based on the sites listed above: Lehringen, (Turner, 2000 - this issue). Weighing all arguments Grobern, Grabschutz, Neumark-Nord, Rabutz, the author assumes that they are Eemian faunas; a Schonfeld, Taubach, Burgtonna and Stuttgart-Un conclusion that implies that Apodemus maastrichtiensis terturkheim. The correlation of these faunas with the was also extant during the Eemian stage. Eemian, based on the lithostratigraphical position of the deposits (overlying Saalian till) or radiometric da Referring Neumark-Nord, Rabutz and Grabschutz ta, seems to be well-established. There is, however, as to the Eemian does not imply that the author neglects mentioned above, still an ongoing discussion about the arguments proposed by Mania (1997) and Mai the age of a number of these faunas. Von Koenigswald (1990 a,b,c). These authors point to differences be-

278 Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 tween the mollusc associations and flora from these more than the areas to the west or to the north of the localities and the mollusc and flora data from the basin influenced by the presence of mountains in the 'classical' Eemian sites Taubach and Burgtonna. Dif west/southwest of the basin (Kahlke, 1990). This ferences in associations do, however, not explicitly might explain the presence of steppe dwellers in the imply dating to different interglacial phases, i.e. inter faunas from the eastern part of Germany. There are, vals separated from each other in time by a glacial however, also steppe dwellers {Stephanorhinus hemi phase. One option might be that there were different toechus, Equus hydruntinus) in the fauna from Stutt- episodes within the same interglacial or with geo gart-Unterturkheim, indicating a mosaic environment graphical variation. Hardly anything is known cur in the area. rently about the faunal changes during the Eemian. Sequences with a good record of molluscs are known General conclusions from, among other places, Burgtonna (Mania, 1973), but mammalian sequences from the Eemian showing In summary, the knowledge of the mammalian fauna the mammalian changes within the Eemian intergla from the Eemian interglacial is based on a restricted cial are so far poorly known. At Neumark-Nord, Gro- number of faunas and is still far from complete. A de bern and Grabschutz several fossiliferous layers were tailed picture of the faunal evolution is unavailable. exposed, representing different episodes within the Only a general picture of the Eemian fauna of north Eemian. A detailed picture of the faunal evolution is western and central Europe can be reconstructed ba missing, however. We only have a general picture of sed on the record of the climatic optimum of the the Eemian fauna of north-western and central Euro Eemian interglacial. Taking into account that the pe, based on the record from the localities mentioned interglacial represented only a very short episode in above. The faunas date from the climatic optimum of the earth history, the number of rather well-dated the Eemian interglacial (Speleers, unpubl.). Eemian faunas is, however, remarkably large. The fauna is well-documented for the eastern part of Environmental conclusions Germany but the Eemian s.s. fossil record is poor in other areas. Despite this, geographical differences in The ecological demands of the European pond tor faunal composition are apparent; Hippopotamus oc toise, Emys orbicularis, the hippopotamus, Hippopota curred in the Rhine valley together with the water mus amphibius, and the water buffalo, Bubalus murren- buffalo {Bubalus murrensis); both species are absent in sis, present in the Eemian fauna, indicate a climate central Germany. In the future, investigations should with relatively high summer temperatures (mean July be focused more on the faunal evolution within the temperature > 18°C) and the absence of periods of Eemian. The results of these investigations will con severe or long-lasting winter frost (Van Kolfschoten, tribute to the debate on the stratigraphical position of 1995). The mammalian fauna of central Europe - the poorly dated faunas and to a better understanding with several deer species as well as the wild boar {Sus of environmental changes, in the past as well as in the scrofa), the straight-tusked elephant {Palaeoloxodon future. antiquus), the browser, Stephanorhinus kirchbergensis, and the dormouse, Glis glis - also indicate the presen Acknowledgements ce of a forested environment. These species occurred together with species indicative of more open envi The author would like to thank Dr. L. Maul (Wei ronment, such as Cricetus cricetus, Equus sp. (or Equus mar) and Dr. R.-D Kahlke (Weimar) for the inspiring taubachensis), Equus hydruntinus and Stephanorhinus discussions and their comments on the first draft of hemitoechus. In summary it can be stated that the the manuscript. Dr. P.L. Gibbard and Ms S.Yates im mammalian record of Central Europe indicates a mo proved the English text, for which I am grateful. saic environment with an alternation of forested and more open vegetation: a picture that contradicts that References of the palaeo-environmental reconstructions based on the palaeobotanical record (Roebroeks et al., 1992). Bastin, B., 1992. Analyse pollinique des sediments detritiques, des Palaeobotanists assume an oceanic climate in western coprolithes et des concretions stalagmitiques du site prehistori- and central Europe during the climatic optimum, que de la grotte Scladina (Province de Namur, Belgique). Etu des et Recherches Archeologiques de l'Universite de Liege 27: with uniform deciduous forest vegetation (Zagwijn, 59-77. 1989). Local conditions might explain the discrepan Benecke, N., Bohme, G. & Heinrich, W.-D., 1990. Wirbeltierreste cy. The climate in the Toringian basin and the sur aus interglazialen Beckensedimenten von Grobern (Kr. Grafen- rounding areas is, nowadays as well as in the past, hainichen) und Grabschutz (Kr. Delitzsch). Altenburger Natur-

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 279

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.168, on 05 Oct 2021 at 10:51:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0016774600021752 wissenschaftliche Forschungen 5: 231-281. Kahlke, R.-D., 1990. Der Saiga-Fund von Pahren. Ein Beitrag zur Bocherens, H., Billiou, D. & Mariotti, A., 1999. Palaeoenviron- Kenntnis der palaarktischen Verbreitungsgeschichte der Gattung mental and palaeodietary implications of isotopic biochemistry Saiga GRAY 1843 unter besonderer Berucksichtigung des Ge- of last interglacial neanderthal and mammal bones in Scladina bietes der DDR. Eiszeitalter und Gegenwart 40: 20-37. Cave (Belgium). Journal of Archaeological Science 26: 599-607. Kahlke, R.-D., 1994. Taubach, Gebiet der ehemaligen Travertin- Bohme, G., 1989. Die Amphibien- und Reptilienreste der Fund- bruche. Altenburger Naturwissenschaftliche Forschungen 7: stelle Bilzingsleben. Ethnographisch-Archaologisch Zeitschrift 366-367. 30: 370-378. Litt, T.j 1990. Pollenanalytische Untersuchungen zur Vegetations- Bohme, G., 1991. Amphibien- und Reptilienreste aus dem Eem-in- und Klimaentwicklung wahrend des Jungpleistozans in den Bee- terglazial von Schonfeld, Kr. Calau (Niederlausitz). In: Striegler, ken von Grobern und Grabschutz. Altenburger Naturwissen S. & Striegler, U. (eds.): Eem von Schonfeld 1. Sonderheft Natur schaftliche Forschungen 5: 92-105. und Landschaft in der Niederlausitz NaturwissenschaftlicherVe- Litt, T, 1994. Palaookologie, Palaobotanik und Stratigrafie des rein der Niederlausitz, Cottbus): 117-129. Jungquartars im nordmittel europaischenTiefland: unter beson Bratlund, B., 1999. Anthropogenic factors in the thanatocoenose of derer Berucksichtigung des Elbe-Saale-Gebietes. J. Cramer (Ber the last interglacial travertines at Taubach (Germany). In: The lin-Stuttgart): 185 pp. role of early humans in the accumulation of Lower and Middle Litt, T & Turner, C, 1993. Arbeitsergebnisse der Subkommission Palaeolithic bone assemblages. Romisch-Germanisches Zentral- fur Europaische Quartarstratigraphie: Die Saalesequenz in der museum Mainz Monographien 42: 255-262. Typusregion (Berichte der SEQS 10). Eiszeitalter und Gegen Brunnacker, K., Jager, K.D., Hennig, G.J., Preuss, J. & Griin, R., wart 43: 125-128. 1983. Radometrische Untersuchungen zur Datierung mitteleu- Mai, D.H., 1990a. Die Flora des Interglazials von Grabschutz ropaischer Travertinvorkommen. Ethnographisch-Archaologisch (Kreis Delitzsch). Altenburger Naturwissenschaftliche For Zeitschrift 24: 217-266. schungen 5: 116-137. Conard, N.J., Preuss, J., Langohr, R., Haesaerts, P., Van Kolfscho- Mai, D.H., 1990b. Die Flora des Interglazials von Grobern (Kreis ten, T, Becze-deak, J. & Rebholz, A., 1995. New geological re Grafenhainichen). Altenburger Naturwissenschaftliche For search at the Middle Paleolithic locality of Wallertheim in Rhein- schungen 5: 106-115. hessen. Archaologische Korrespondenzblatt 25: 1-27. Mai, D.H., 1990c. Zur Flora des Interglazials von Neumark-Nord, Eissmann, L., 1990. Das mitteleuropaische Umfeld der Eemvor- Kreis Merseburg. Veroffentlichungen Landesmuseum fur Vorge kommen des Saale-Elbe-Gebietes. In: Eissmann, L. (ed.): Die schichte Halle 43: 159-160. Eemwarmzeit und die fruhe Weichseleiszeit im Saale-Elbe-Ge- Mania, D., 1973. Palaookologie, Faunenentwicklung und Strati- biet. Altenburger Naturwissenschaftliche Forschungen 5: 11-48. graphie des Eiszeitalters im mittleren Elbe-Saalegebiet auf Eissman, L., Litt,T, Weber, T, Hentschel, D.,Wimmer, R. &Wan- Grund von Molluskengesellschaften. Geologie 21: 1-175. sa, S., 1988. Bemerkenswerte Eemvorkommen in glaziaren Bee- Mania, D., 1992. Neumark-Nord ein fossilreiches Interglazial im ken der Saaleeiszeit sudlich der "Warthe-Linie" im Raum Halle- Geiseltal. Cranium 9 (2): 53-76. Leipzig. Zeitschrift fur GeologischeWissenschaften 16: 543-555. Mania, D., 1997. Das Quartar des Saalegebietes und des Harzvor- Erd, K., 1990. Pollenstratigraphie des interglazialen Beckentons landes unter besonderer Berucksichtigung der Travertine von von Rabutz sudostlich Halle/Saale. Altenburger Naturwissen Bilzingsleben - Ein Beitrag zur zyklischen Gliederung des eura- schaftliche Forschungen 5: 141-147. sischen Quartars. In: Mania. D. (ed.): Bilzingsleben 5. Verlag Erfurt, J. & Mania, D., 1990. Zur Palaontologie des jungpleistoza- Ausbildung +Wissen (Bad Homburg/ Leipzig): 23-105. nen Waldelefanten von Grobern, Kreis Grafenhainichen. Ver Mania, D., 1999. 125.000 Jahre Klima- und Umweltentwicklung offentlichungen des Landesmuseums fur Vorgeschichte in Halle im mittleren Elbe-Saale-Gebiet. Hercynia Neue Folge 32: 1-97. 43:215-224. Maul, L., 1994. Erster Nachweis von Hystrix in der Pleistozan- Fischer, K., 1991.V6gel und Grofisaugetiere aus dem EemTnter- Fundstelle Burgtonna (Thuringen, Mitteldeutschland). Sauge- glazial von Schonfeld, Kr. Calau (Niederlausitz). In: Striegler, S. tierkundliche Information 18: 673-682. & Striegler, U. (eds.): Eem von Schonfeld 1. Sonderheft Natur Pfeiffer, Th., 1998. Die fossilen Damhirsche von Neumark-Nord und Landschaft in der Niederlausitz NaturwissenschaftlicherVe- (Sachsen-Anhalt). Eiszeitalter und Gegenwart 48: 72-86. rein der Niederlausitz, Cottbus): 169-176. Pfeiffer, Th., 1999. Alces latifrons (Johnson 1874) (Cervidae, Mam Griin, R., Brunnacker, U. & Hennig, G.J., 1982. 230Th/234U-Da- malia) aus den jungpleistozanen Kiesen der Oberrheinebene. ten mittel- und jungpleistozaner Travertine im Raum Stuttgart. Neues Jahrbuch fur Geologische und Palaontologische Abhand- Jahresbericht und Mitteilungen der oberrheinische geologische lungen 211: 291-327. Verein 64: 201-211. Roebroeks, W, Conard, N.J. & Van Kolfschoten, Th., 1992. Dense Hartung, W., 1991. Die Entdeckung des Waldelephantenschlacht- forests, cold steppes, and the Palaeolithic settlement of northern platzes von Grobern, Kr. Grafenhainichen. Veroffentlichungen Europe. Current Anthropology 33: 551-581. des Museums fur Ur- und Fruhgeschichte Potsdam 25: 101-103. Schweigert, G., 1991. Die Flora der Eem-interglazialen Travertine Heinrich, W.-D., 1982. Zur Evolution und Biostratigraphie von Ar- von Stuttgart-Unterturkheim (Baden-Wiirttemberg). Stuttgarter vicola (Rodentia, Mammalia) im Pleistozan Europas. Zeitschrift Beitrage zur Naturkunde, Serie B 170: 73-94. fur GeologischeWissenschaften 10: 683-735. Soergel, W, 1920. Der rabutzer Beckenton, Geologie, Palaontolo Heinrich, W.-D., 1991. Die Kleinsaugerfauna des Eem-Interglazials gie, Biologic Veroffentlichungen des Provinzialmuseums zu Hal von Schonfeld, Kr. Calau (Niederlausitz). In: Striegler, S. & le 1 (4): 1-40. Striegler, U. (eds.): Eem von Schonfeld 1. Sonderheft Natur und Steiner,W., 1976. Das geologische Profil des Travertin-Komplexes Landschaft in der Niederlausitz Naturwissenschaftlicher Verein von Taubach bei Weimar. Quartarpalaontologie 2: 83-118. der Niederlausitz, Cottbus): 177-188. Toepfer,V, 1958. Steingerate und Palokologie der mittelpalaolithi- Kahlke, H.D., 1977. Die Rhinocerotidenreste aus den Travertinen schen Fundstelle Rabutz bei Halle (Saale). Jahresschrift fur Mit- von Taubach. Quartarpalaontologie 2: 305-359. teldeutsche Vorgeschichte 41/42: 140-177. Kahlke, H.-D., 1978. Das Pleistozan von Burgtonna inThuringen. Turner, C, 2000. The Eemian interglacial in the North European Quartarpalaontologie 3: 399. plain and adjacent areas. In: Van Kolfschoten, Th. & Gibbard,

280 Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000

Geologie en Mijnbouw / Netherlands Journal of Geosciences 79(2/3) 2000 281