Recycling of Badger/Fox Burrows in Late Pleistocene Loess by Hyenas at the Den Site Bad Wildungen-Biedensteg (NW, Germany): Woolly Rhinoceros Killers and Scavengers in a Mammoth Steppe Environment of Europe

The Late Pleistocene (MIS 5c-d) Ice Age spotted hyena open air den and bone accumulation site BadWildungen-Biedensteg (Hesse, NW, Germany) represents the first open air loess fox/badger den site in Europe, which must have been recycled by Crocuta crocuta spelaea (Goldfuss, 1823) as a birthing den. Badger and fox remains, plus remains of their prey (mainly hare), have been found within the loess. Hyena remains from that site include parts of cub skeletons which represent 10% of the megafauna bones. Also a commuting den area existed, which was well marked by hyena faecal pellets. Most of the hyena prey bones expose crack, bite, and nibblingmarks, especially themost common bones, the woolly rhinocerosCoelodonta antiquitatis (NISP = 32%).The large amount of woolly rhinoceros bones indicate hunting/scavenging specializing on this large prey by hyenas. Other important mammoth steppe hyena prey remains are fromMammuthus primigenius, Equus caballus przewalskii, Bison/Bos, Megaloceros giganteus, Cervus elaphus, andRangifer tarandus.The fewdamaged bone remains of a scavenged cave bearUrsus spelaeus subsp. are unique for an open air situation. Abundant micromammal, frog, and some fish remains were concentrated in “pellets” that contain mainly mammoth steppe micromammals and also frog and fish remains that seem to originate from the nearby river/lake.

Few contemporary used hyena and Neanderthal sites have been described from hyena dens in mammoth steppe lowlands and adjacent cave-rich region environments of north-central Europe, in England and Germany [9,16]. The degree of prey bone damage and presence/absence of "nibbling sticks" and faecal pellets or hyena population structure and their individual amount allow the reconstruction, much better, of the ethology of the last hyenas of Europe. The discussions for nonarchaeological sites no longer focus only on the human/carnivore origin discussion. Although hyena cave-den sites predominate in the European fossil record (e.g., Germany in [17]), open air sites may have been much more common throughout the mammoth steppe lowlands of Europe, but have been overlooked or not identified as such (cf. Westeregeln or Bottrop sites in [10,18]).
Open air hyena den sites in loess deposits without human impact are not analyzed in Germany, as yet, whereas other bone accumulation sites on river terraces have been analyzed along the Emscher River near Bottrop in the Westphalian mammoth steppe lowland [10]. Recently many open air hyena den sites (loess, gypsum karst, river terraces: Saalfeld, Bottrop, Westeregeln, Sewecken-Berge, Thiede, and others) from Germany have been described [17,[19][20][21], whose density     summer times, when the permafrost soil was soft in the upper parts.
The bioturbation interpretation would fit into the "hyena commuting/prey storage site, " but can no longer be studied because of the nonopen loess pit Biedensteg. In this section ( Figure 2) such depressions are figured as hyena prey depots. Possibly, a later cryoturbation, a result of permafrost soils fitting into the environment and climatic situation of that time, was responsible for secondary overprint of the primary sediment structures. Bioturbation by mammoths on lake shores, which left depressions of their footprints, must be taken into account, as is discussed for other sites (cf. [37]).
The "pellet horizon" is figured differently in the publications (of Jacobshagen et al., 1963, [32]). The section of Kulick [32] indicates that the pellets and the macromammal bones are mixed in a single horizon. Proof for that might be caliche concretions around hyena coprolites in which micromammal bones and teeth are also cemented in. The "hyena prey depot site" and the "pellet horizon" are from the same period and are dated relatively (no absolute data) into the late Middle Late Pleistocene or Weichselian (65.000-90.000 BP, MIS 5cd, Figure 2).
The bone-rich horizon is overlain by another palaeosoil, the "Lohner Soil, " which can be found in the region at different sections [31,32]. After their interpretations a solifluction of Loess and Wilde river gravel material took place in the middle Late Pleistocene warm period ( Figure 2). V. vulpes and M. meles were the dominating faunal elements, besides L. europaeus. This fauna fits to Meles/Vulpes den burrow sites in loess soils, in front of which they often left some prey bones.
Finally the upper loess was deposited within the LGM, and after, the upper part was decalcified during the Holocene period. The "Eltviller Tuff " is a one to two centimeter thin layer in the upper loess and the only absolute dated horizon with an age of around 16.000 BP ( [31], Figure 1(c)).

Small Carnivore Fox Den and Mustelid Bone Assemblage
Meles meles (Linné 1758) (Figure 4(13)-(32)) ( Table 5) Table 1). Vulpes vulpes (Linné 1758) (Figure 4(1)-(9)) remains consist of 13 common fox bones (Table 3) including a skull. This skull is incomplete, as most of the anterior part with its dentition is missing. The last three teeth are in the left maxillary ( Figure 4(1)). From a right forelimb the scapula, humerus, and radius were found, which seem to belong to one individual (Figure 4(2)-(4)). From a hind limb, not only the left femur shaft and incomplete tibia but also a right calcaneus and a metatarsus III are represented (Figure 4(5)-(8)). A fragment of a metapodial is missing its proximal joint. Finally a lumbar vertebra and one rib are preserved. The pelvis is missing its left part (Figure 4(9)). A second pelvis fragment is again incomplete. Material from two individuals is present, indicated by the pelvis remains. Possibly most of the bones belong to only one individual. All postcranial bones show a complete fuse of the symphyses and are from either a single animal or several adult animals.
Vulpes lagopus (Linné 1758) (Figure 4(10)-(12)) ( Table 4) was found with a nearly complete skull, without the jugal arches, but with the right mandible (Figure 4(10)- (11)). The skull sutures are not fully fused and teeth are barely used; therefore it was a young adult individual, as only a single individual can be estimated from the bone material. The postcranial material is present with a femur shaft and pelvic fragment (Figure 4(12)).
Mustela putorius Linnaeus 1758 (Figure 4(33)) ( Table 6) is present with a single half skull (Figure 4(33)) of which the anterior part with most of the dentition is preserved.
Lepus europaeus/timidus Linné 1758 ( Figure 14(1)-(9)) ( Table 14) is represented by 28 bones which are cranial fragments, two are mandibles and the rest are postcranial bones (Table 13). There is an articulated pedal skeleton ( Figure 14 (9)) and an articulated pelvis with lumbar vertebral column ( Figure 14 (5)). The figured material ( Figure 14) seems to be from one individual, which is indicated by the bone preservation and articulations. Another argument is the individual adult's age and the fresh fractures of the humerus, radius, the right femur and left tibia, or some processes of the vertebrae, which were caused during the excavations. Bones from other individuals of young and adult age are also preserved and have been completely disarticulated. 25% of the remains are from young animals; 75% are from adult hares. Three animals can be estimated by the tibia as minimum individual number.

The Hyena Population and Coprolite Remains
The Ice Age spotted hyena Crocuta crocuta spelaea [1] ( Figure 2) skeletal remains consist of four skulls, three mandibles, one radius, and a femur (Table 1). Additionally, there are 16 coprolites which were rescued. From the first skull ( Figure 2(1)) deformations do not allow exact metric data. The second skull (Figure 2(2)) is 290 mm in total length and measures 265 mm between the incisive and condyle. The largest height is behind the frontal processes (114 mm). The distances between the canines and P 4 are about 68 mm. The width of the frontals (zygomatic processes) measures 90 mm. Finally the outer distance between the canines is 58 mm. The largest diameter of the canines in the middle of the tooth is 18 mm. The brain case symphyse of the third animal (Figure 2(3)) is slightly fused and articulated. The parietal, frontal, palatine, and temporal are incomplete. The maximum width measured, between the temporal, 73 mm, whereas it is preserved in 76 mm in length.
One left mandible (Figure 2(4)) is of an adult animal and might belong to one of both individual adult skulls, which show a similar tooth use stage. The jaw was cracked by hyenas between the P 2 and P 3 ; the P 3-4 and M 1 are present. The ramus was damaged during excavations.
A few postcranial bones are represented with one axis of an adult animal exposing bite damage marks (Figure 2(6)). A left radius and a left femur (Figure 2(5) and (7)) are from one very young cub, both being incomplete as a result of scavenging activities by large carnivores.
Coprolite Material. The hyena coprolites are generally white inside and the pores are filled with iron and manganese minerals. The coprolites show a moderate variability and even bone contents (Figure 2 (15)). They can point to both sides or can end round to flat on one side as a result of attachment to another pellet. Other pellets are "unshaped" and irregular. These were often found in the non-spindlelike pellet aggregations (Figure 2(10)). In the material from Biedensteg each coprolite contains several bone fragments, which are often visible on the surfaces (Figure 2(11)-(12)). These are small pieces, well rounded by stomach acid, and are mainly from the bone compacta, but also are isolated pieces of bone spongiosa. This spongiosa is very thin walled and should have been completely dissolute. These spongiosa pieces are most comparable to the bone spongiosa of the woolly rhinoceros, but might also refer to other megamammals.

Hyena Megafauna Prey Remains
Ursus spelaeus Rosenmüller 1794 subsp. (Figure 3) is represented by four cave bear bones and fragments. The left scapula (Tables 2 and 3(1)), which lacks all distal parts seems to be destroyed by hyenas. Large carnivore gnawing and bite marks are visible at the glenoid. A right humerus shaft (Figure 3(2)) is missing the joints as a result of heavy carnivore chewing. At the shaft ends and in the lower middle, bite marks are present. The diameter of the bone shaft is small, being only 49 mm. From one left incomplete ulna (Figure 3(3)) the distal joints were chewed and also some bite marks are visible. The 50 mm maximum width ulna has, again, small proportions.
Finally, a fragment of a femur shaft (Figure 3(4)) with heavy chewing damage indicate the cracking and further use of the bone fragment as a typical hyena "nibbling stick" (for teething purposes of hyena cubs).
Mammuthus primigenius (Blumenbach 1799) ( Figure 12 (1)-(3)) is represented by three remains consisting of a tooth lamella fragment from a juvenile animal, a thoracic vertebra neural arch and centrum fragment, and a long bone fragment used as a nibbling stick ( Table 7). The material is from adolescent elephants.
Coelodonta antiquitatis (Blumenbach 1799) ( Figures 5-11) is the most abundant, listed in Table 8. The cranial elements consist of a middle part of a skull from a young calf ( Figure 9).    The connection in-between the maxillas were restored in former times. Originally, the maxillary part between the teeth was damaged by hyenas. All three dm 1−3 milk teeth on both sides are present ( Figure 6(1a)-(1d)). Both m 1 's are breaking through, whereas the m 2 's were still in the maxillary.
These are not present, but the alveolar grooves are preserved. This skull was badly damaged by the hyenas, especially at the anterior part and the brain case. The latter shows a very interesting large carnivore brain case opening. There are some bite marks, but thin parallel long scratch  Both mandibles of the lower jaw ( Figure 6(1e)-(1h)) fit to the skull by the identical milk dentition of the dm 1−3 and the tooth rising of the m 1 . Both jaws were cracked in the symphyses area and have old fractures. Additionally, they are lacking the rami and have large carnivore chewing and gnawing marks ( Figure 6(1e)-(1h)). The left jaw possesses the dm 1−3 and the m 1 . The right mandible was damaged by the excavations and because of this is lacking the anterior part, including the dm 1-2 . Other cranial material was described and partly refigured by Jacobshagen [34]. He refigured some lower jaw teeth of one individual (right P 3-4 , M 1 , and left M 2-3 ). The little use of the M 3 indicates an origin of an early adult animal. It is suggested here that these belonged most probably to the skeleton of an early adult female individual ( Figure 5(b)). Scapulae are preserved with one nearly complete left shoulder blade (Figure 7(1)). Some parts from the left side and joint area, destroyed by the excavations, were restored. Bite marks were found only distally. Here, hyenas left typical chewing marks in the very soft scapula. The margin is therefore typically irregular, resulting from cracked bone material. The scapula seemed to belong to the female skeleton. A second fragment of a scapula is Chewed (4b) 10 cm in preservation and could be found in a lower horizon. One humerus is described by Jacobshagen [34], which can no longer be located. It was a right humerus that was chewed on the proximal joint. Ulnae are present with five bones (Figure 8(1)-(4)) from different old animals. The most juvenile, a neonate to young, animal's left ulna must have been articulated to one radius ( Figure 6(2)). This result is from the comparison to an articulated right ulna/radius from a young adult to adult animal whose joints are chewed away in the same way (Figure 8(1)). The latter might belong to the young adult female rhinoceros ( Figure 5(b)), of which also other bones were found partly articulated. At least seven radii (Figure 7(4)-(6), MNI = 7) were found, of which four are from young adult to adult animals and the last from the neonate to very young individual. The four pelvis remains are typical rests of hyena feeding activities ( Figure 10(1)- (3)). The acetabular and surrounding two acetabular fragments are from different animals. The one figured (Figure 10(1)) has not only hyena, but also arctic fox, wolf or hyena cub, and even small rodent nibbling marks. The fourth pelvis remain is only a part of the ileum (Figure 6(3)) and seems to belong to the juvenile animal, because it is also chewed from the acetabular region. It is also heavily chewed at the soft distal part with irregular margin. Four femora are preserved, of which one is a fragment, a second is from a juvenile animal ( Figure 6(4)), and a third and fourth are from an adult C. antiquitatis (Figure 10(4)- (5)). Another fragment is of an adolescent, with strong chewing marks (Figure 10(6)).
As described by Jacobshagen [34], there was a right femur ( Figure 10(4)) found in articulation with a tibia (Figure 11(2)). Only one nearly complete left patella (Figure 11(9)) was excavated and might belong also to the female skeleton's hind leg ( Figure 5(b)). The tibia has very typical hyena caused damages and is in an early stage (stage 1) of destruction. Also this fits well with the partly articulated female skeleton carcass. Three tibiae are very massive and have a strong width in the shaft (Figure 11(3)- (5)). All tibiae compared indicate a sexual dimorphism with males being stronger and more massive in their bones. Mostly the proximal joint was chewed away first, although at the distal part in a middle stage (stage 2 of three) of bone feeding, two groves were left, which is documented at all three tibiae (Figure 11(3)- (5)). Two fibula 10 cm    remains are in the material, with one ( Figure 11 (7)) being proximally incomplete as a result of the excavations. That one was articulated to one tibia in the stage of hyena chewing and seems to belong to the female carcass ( Figure 5(b)). The distal part shows long bite scratches. The second fibula was cracked away from a tibia and was left with the middle shaft with bite marks at both ends ( Figure 11(6)). Only one astragalus and calcaneus are in the material (Figure 11 (8)) also most probably belonging to the hind leg of the female skeleton ( Figure 5(b)). They fit perfectly together, indicated additionally by overlapping bite scratch marks which are crossing both bones. After the descriptions by Jacobshagen [34] there were three complete metatarsals (2-4) that also fit for the female skeleton ( Figure 5(b)), although it is unclear whether they are from the right or left side. All vertebrae show the typical hyena chewing by the lack of nearly all processes.
They seem to be all from one nearly adult individual, indicated by a series of articulation and the similar degree of nonfusing of the caudal vertebra centrum disc. The cranial disc, in contrast, is already fused completely at all vertebrae. From the vertebral column, the first three cervical vertebrae were found connected (Figure 9(4)). Atlas (Figure 9(1)), axes (Figure 9(2)), and the third cervical vertebra (Figure 9(3)) have bite marks on the damaged processes. The next articulated vertebral column part is the vertebra from the sixth cervical to the first thoracic (Figure 9(7)). Articulated cervical vertebrae no. 6 ( Figure 9(5)) and no. 7 (Figure 9(6)) and thoracic vertebra no. 1 (Figure 9(8)) are also lacking most of their processes, especially the dorsal ones. Two more articulated vertebrae are the second (Figure 9(9)) and third (Figure 9(11)) thoracic vertebrae which are heavily chewed (Figure 9(12)). The fourth thoracic vertebra (Figure 9(10)) was only a centrum that was found in nonarticulation with other vertebrae. The complete neural arch was eaten. Parts of the left side were cut by excavation activities. The longest articulated vertebral column part exists from the sixth to ninth thoracic vertebrae (Figure 9(17)). Typical for the hyena scavenging activities are the chewed dorsal spines. Finally, the articulated last thoracic and first lumbar vertebra were found connected (Figure 9(18)). Also, the first lumbar vertebra is lacking parts of the proc. transversus. The ribs generally have no hyena bite marks, but obviously they were removed from the carcass (Figure 9(20)- (28)). All costae have cracking fractures at both ends; all joints are lacking. Only one small rib fragment (Figure 9(28)) has distally small bite marks.
Nibbling by a small carnivore, such as a young hyena, wolf, or arctic fox, has caused a pointed distal end. A small fragment was used for nibbling by young hyenas ("nibbling stick" no. 3, Figure 9(27)). The present rib fragments are from the anterior part around the forelimb, and a few are from the last thoracic vertebrae.
Bison/Bos (Figure 12(4)-(9)) remains consist of 13 bones ( Table 9), two of which are teeth, the others being postcranial bones, which are all incomplete as a result of large carnivore activities. Most bones are limb bones, especially from the hind limbs. The teeth are two M1's, one from the upper and the other from the lower jaw. The strong tooth use indicates an individual of adult to older adult age. From the forelimb a metacarpal fragment (Figure 12(4)) was found. The metacarpal shows a typical hyena cracking preservation; the distal part has sharp edges. Most bones are from the hind limbs. Both femora were cracked in the middle of the shaft but also the distal joints were heavily eaten and nibbled ( Figure 12(5)-(6)). One middle shaft of a cracked tibia (2) 10 cm and one proximally chewed calcaneus (Figure 12(8)) and two femur fragments seem to originate of the rigth hind limb of one animal. Finally, there is one thoracic vertebra centrum (Figure 12(9)) and one cervical vertebra (Figure 12(10)). The processes were chewed, and also some deep scratch bite marks can be found ventrally. All bones belonged to one, or possibly a few adult individuals. Equus caballus przewalskii Poljakoff 1881 (Figure 13(4)-(15)) consists of 19 bones, of which two are mandible fragments, one cranial fragment and a single tooth, although mainly leg remains are represented (Table 10). The one metacarpus is 236 mm in length and distally 50 mm in width (Figure 13(8)) and falls within the small Przewalskii horse metapodial osteometry (cf. [9-11, 18, 37-44]). The same is for one complete metatarsus (Figure 13 (15)) with its 257 mm length and 53 mm distal width. Also, there is the nearly complete lower jaw of a male horse (Figure 13(4)), as well as other small-sized bones from the smaller Przewalskii horse. There are bones from young horses (21%), with all others being from adult individuals (79%).
Megaloceros giganteus (Blumenbach 1799) (Figure 13(1)) was found with only seven bones, including one mandible fragment and three teeth, all from adult animals ( Table 11). The material described and figured from Jacobshagen [34] is lost.
Cervus elaphus Linné 1758 (Figure 13(2)- (3)) is present with only two remains (Table 13). From the cranium, a right maxillary fragment with two M 1-2 shows the M 2 not in a developed state, although, the M 3 alveolar is opened and the tooth is in change. Another remain is a metatarsus (3) 10 cm     Table 12) is more common, with 24 remains. The rest of the bone material, such as a right metatarsus, a phalanx 1 and phalanx 2 proximal joint disc, and a right radius distal joint fit in the nonfusing of the joints to one young animal. The dropped antlers are from males and are all from sheds, which must have been collected by hyenas. Similar damages are present on the distal ends where large triangular-oval bite impact marks and elongated scratches indicate large carnivore damage (Figure 13(15)-(17)).

The Badger/Fox Types and Den Micromammals and Pellet
Accumulators. At open air badger den sites, typically, most skulls and massive long bones were found, although such long-term used badger loess den systems are described [45]. In those, bone accumulations are dominated by skull remains, being figured, for example, for the Schneehalle Cave (South Germany, [46]). Commonly, badgers die in their dens [46][47][48], explaining their bone accumulations in burrows and caves. The amount of bones, mainly of senile and very young badgers of Bad Wildungen, fit into such a scheme. Bite marks  and missing joints in a humerus and tibia might be the result of badger cannibalism [47] or even hyena activities. The skull and postcranial material can be referred to the Asian species Meles meles cf. leucurus (cf. [49,50]), and the skull seems to be of male origin (cranial sexual dimorphism; see [51]). This is so far important, because this subspecies seem to have immigrated to Europe from Asia during the Late Pleistocene, where it is nowaday's extinct [50]. The badger, with its diet (cf. [52]), was not responsible for the bone accumulations of medium-sized mammals and anures, or reptiles, but of micromammals (cf. [53]), also at the Bad Wildungen-Biedensteg open sir site. Foxes (V. lagopus and V. vulpes) might have reused the badger burrows [48]. Fox bones and skulls are typically found at those fox den sites and would explain, additionally, the presence of smaller mammal fox prey remains, especially hares and the micromammal pellets generally found at modern fox dens (cf. [48]).
Quaternary small mustelids in central Europe are rare in the fossil record outside caves (cf. [54,55]). Their pellets can contain anure or fish bones. Frog or fish remains from Bad Wildungen seem to be partly of prey deposits of Mustela putorius. The small marten type is storing along small rivers or lakes, fishes, frogs, and other animals [48].
A especially high amount of frog bones must have resulted, additionally, from other large water birds and/or other predators which also left pellets and bone remains at the river and along the lake.

Hyena Population and Cannibalism.
The hyena skulls from Bad Wildungen-Biedensteg are from female hyenas which are similar to many other skulls of central Europe (cf. [17]) and are anatomically interesting in their dentition (partly absence of M 1 ), but fall into the variability of C. c. spelaea. A brain case, two incomplete limb bone shafts, a left radius, and a left femur are fitting for a single cub, which are very small in their proportions. They also have bite marks and must have been chewed, as compared to other cannibalistic damaged hyena long bone finds from Europeans caves (cf. [11,22,23,25,27,56]). Their proportions fit best for a very young cub, maybe only of a few days or weeks in age, compared to the cub material from the Srbsko-Chlum-Komin Cave, Czech Republic [11]. The young hyena was possibly eaten cannibalistically, possibly by another cub, due to competition (cf. modern African hyenas in [57][58][59]). All bones of the Bad Wildungen hyena population and even the skulls have nibbling, chewing, and cracking marks of hyenas. The lack of the jugals and temporal parts of the skulls is the result of cracking the lower jaws from their joints, which is demonstrated for many skull finds in Europe (cf. [17]). The scavenging of their own species leaves dominantly cranial remains at not only den sites, such as the skulls, lower jaws, and teeth, but also the long bones (e.g., modern spotted hyenas, [60,61]). Scavenging of their own is best documented in the Srbsko-Chlum-Komin Cave [11]. The dominance of cranial material at Bad Wildungen hyena den site is comparable not only to the German Perick Caves and Rösenbeck Cave and other Sauerland Karst hyena dens, but also to other caves, such as the Czech Sloup Cave, Výpustek, in the Bohemian and Moravian Karst regions [7,17,56]. Vertebrae and rib bones are underrepresented at most hyena den sites (especially at birthing dens and prey storage den types), the exceptions being where complete articulated skeletons are found at prey storage sites, such as were found
The large bone enrichment at Bad Wildungen was already identified as a product of the activities of C. c. spelaea [35]. The comparison of different Late Pleistocene C. c. spelaea hyena cave and open air den sites in Europe allows a classification of the den type, by separating three main age classes: (1) cubs, (2) adolescents, and (3) adult-senile individuals ( Figure 15). The high presence of cubs indicates, similarly as in modern spotted hyenas [57,[67][68][69], birthing dens. Other

22
Journal of Geological Research  indicators for such birthing dens are "nibbling sticks. " At Bad Wildungen there are three such chewed bone fragments: one of a mammoth, whose bone fragments are found at birthing dens [70] for teething purposes of hyena cubs [7]; the other nibbling sticks are from Coelodonta and Ursus bone fragments. These birthing dens are generally recycled from medium-sized carnivore, such as porcupines, or by hyenas own excavated burrows, which can be situated nearby commuting dens (cf. modern in [71]). Bad Wildungen must have also been this type of den, where higher amounts of prey remains were accumulated, or even stored (prey storage den type). Similar large bone accumulations at commuting den sites have been reported in Africa from C. c. crocuta (cf. [61,68,[71][72][73][74][75][76][77][78][79][80][81]).   [3, 5-7, 11, 22, 41, 82]. Exact documented excrement markings on a gypsum karst open air den were recently published at the site Westeregeln, Central Germany [9]. A first terminology was published for the pellet shape types [44]. The hyena pellets from Bad Wildungen fall within the hyena pellet shape types. Several smaller pellets are attached to each other, forming spindle-like, or irregular accumulated aggregations, similar to modern African spotted hyena excrements [9]. Modern spotted hyenas are using faecal pellets to mark their territory, especially their den sites [83].  [8,11,65,66]). A high percentage of hyena prey bone remains at the site Bad Wildungen-Biedensteg ( Figure 16) do not represent the real percentages of the prey. It is more demonstrated, for example, at other hyena open air sites, as a result of taphonomy and selection [9]. The bones of the woolly rhinoceros are extremely massive, and, in contrast to nearly all other large 24 Journal of Geological Research   [10]. The open air site Bad Wildungen-Biedensteg has delivered only a very few mammoth bones (2% of the prey bones) which are typical at middle high mountainous hyena dens of Europe, where mammoths seem to have been absent or rare [7]. Hyenas specialized there on cave bear scavenging ( [42], Figure 16). The amount of Przewalski horse remains (8%) is as usual high. In most open air sites and middle mountainous elevated European caves the small Przewalski horse is the main or second dominant prey (up to 50%; [7, 9-11, 18, 37, 40-44]). If all the small carnivores are excluded from the statistics, then the horse remains represent the second largest prey (cf. [85]). Bones of those horses are recorded with small proportioned forms (see metapod discussion) attributed to E. c. przewalskii in Germany or Czech Republic at other hyena den sites of early to middle Late Pleistocene age [7,85]. Late Palaeolithic archaeological sites have the youngest records from the Late Magdalénian [86] or Epipalaeolithic/Early Mesolithic [87]. Finally, trackways have been described from the German Volcanic ashes of the Laacher Volcano to be of Przewalski horse origin [37,88]  art and identified also the horses by the unique "M-sign" (resulting from fur colour and fur change) and "uplifted mane" (only in those horses, not in modern present horses) to represent obviously Przewalski horses within the Late Palaeolithic times (cf. e.g., [86,89]) and especially within the cold periods of the Late Pleistocene.  Figure 16). All bones have medium to massive nibbling, chewing, and gnawing marks, mainly produced by the Ice Age spotted hyenas, as compared to other den sites [10,91] and modern spotted hyenas [92,93]. Scratches deep into the spongiosa of the joints are very typical of hyena origin and can be found at many other European open air and cave sites (e.g., [11, 21-23, 25, 29, 40, 41, 82, 94, 95]). The material from Bad Wildungen consists of a few cranial and mainly postcranial bones of at least five woolly rhinoceros individuals. Remains of a young, less than one-year-old calf, a young adult female, and a few remains of a male adult skeleton can be distinguished ( Figure 5(b)). Besides those, mainly forelimb bones from some other rhinoceros individuals were found. A comparison to a normal bone proportion relation analyses [10] to the material from Bottrop open air site ( Figure 5(a)) shows differences mainly in the thoracic (vertebrae, costae) presence. In Bad Wildungen, those thoracic elements are more abundant, similar to those found on nonscavenged skeletons like the Petershagen skeleton [90], which indicates the scavenging of a carcass very nearby the den.
The presence of a carcass is also demonstrated by the articulated vertebral column ( Figure 5(b)). To this, most probably, other elements belong. An originally articulated right hind limb (femur and tibia, astragalus, and calcaneus) or forelimb bones, such an ulna and radius, support the original presence of one animal carcass which was decomposed in parts. Such decompositions could have taken days, such as what is known for Late Pleistocene elephant carcasses [43]. The carcass of the most probable female C. antiquitatis must have laid on the right side of her body during main carcass feeding activities, because more bones from that side are preserved. The skull is lacking, but it seems as if all isolated teeth found from the lower jaw indicate the complete destruction of the mandibles by the hyenas. Isolated teeth of woolly rhinoceros are typically at hyena den sites (e.g., [10]).  Figure 16: (a) Late Pleistocene spotted hyena sites and dens and woolly rhinoceros remains in NW-Germany. (b) "Cross-section" through the mountain boreal forest cave bear dominated bone assemblages to the mammoth steppe lowland faunal assemblages (composed after [9,10,22,82,90] and new results).
Maybe the skull was cut off by the hyenas or at least destroyed. A few ribs were only cracked, and nearly all are lacking their joints. The long bone joints were not chewed off completely, because of their articulation. This indicates a fresh carcass that was not completely used by the hyenas and was left in an intermediate stage of carcass destruction (cf. Figure 5(a)). After the bone destruction stages, those are in stage 2 sensu Diedrich [10]. The spongiosa remains of woolly rhinoceros were quite often found in the hyena coprolites at the Bad Wildungen-Biedensteg site [35]. The brain case opening of a calf is similarly figured as an adolescent rhinoceros skull from Selm-Ternsche [10], as figured from rhinoceros skull damages from other sites [96]. The finds of juveniles, such as the few-weeks-old rhinoceros (Figures 5(b) and 6), hyena, or the neonate cave bear, fit for the hunting and main activity time of the hyenas at Biedensteg in the late spring and early summer. Other remains of at least four more rhinoceros individuals and other prey remains were imported, possibly from the Ice Age spotted hyenas.

Hyenas as Cave Bear Scavengers.
The cave bear bones might belong to one skeleton of a mature female cave bear [35]. The small diameter, 75 mm, of the scapula glenoid fits for cave bears of the smaller subspecies U. spelaeus subsp. of the early/middle Late Pleistocene, compared, for example, to the cave bear population of the Perick Caves in the Sauerland Karst (Figure 1; [97]) or the newer studied cave bear populations and subspecies of the Rübeland Caves [98]. Also, the other bones and femur fragments were compared to some hundred bones from the Perick and Rübeland Caves, all having again smaller proportions, excluding a U. ingressus cave bear type of the latest Late Pleistocene. Finally, similarly as figured with the "nibbling stick" in the Perick Caves, some cave bear femora and other bone fragment nibbling sticks are present [70], which only hyenas must have produced by teething cubs (cf. [7]). A scavenging of a cave bear carcass outside a cave is the only clear report of such a scenario [97], but is not exceptional, if compared to the hunting/feeding strategies of the Late Pleistocene spotted hyenas. It is now well known that they scavenged cave bear carcasses in the mountain regions of Europe, such as the Sauerland Caves, the Perick Caves, and Rübeland Caves, and additionally several other cave bear dens all over Europe [42,70,98,99].

Conclusion
The open air hyena den site Bad Wildungen-Biedensteg (NW-Germany) must have been located at the margin of an ancient small lake and the Wilde River in a mammoth steppe landscape on the eastern slopes of the Sauerland Mountains during the early to middle glaciation (early late Pleistocene or Weichselian, about "65.000-90.000 BP, " MIS 5c-d). This shallow lake margin, or at least muddy area, was in the center of a large sinkhole structure, which was caused by subsurface dissolution of Zechstein salt in the underground. The sinkhole received freshwater influence by the early Wilde River, indicated by especially freshwater fish remains, but also some other water related animals such as frogs, which were found accumulated in many pellets. Those are excrements of red/arctic foxes, steppe iltis and large carnivore water birds, or owls. Nearby, a badger/fox den burrow area in loess deposits must have been present, where their bone remains and those of their prey (mainly hare, and micromammals) were accumulated, also in pellets. With Biedensteg, an open air hyena birthing and overlapping communal den with prey deposit can be presented with probably reused badger/red fox burrows for the natal den function. 10% of the NISP are Crocuta crocuta spelaea remains, including three grown-up animal skulls, and cranial and postcranial remains of a young cub. Abundant are hyena coprolites (mainly encrusted by caliche), which contain fragments of bones, and most probably quite abundant bone spongiosa fragments from woolly rhinoceros bones. This corresponds to the main hyena prey Coelodonta antiquitatis (NISP = 32%). Another main prey is the horse Equus caballus przewalskii (8%). This dominance of woolly rhinoceros/horses in the Late Pleistocene bone assemblages in northern Europe was caused solely by those large carnivores and is typical of many hyena open air and cave den bone accumulation sites in northern Germany and Czech Republic (central Europe). a newer house construction, where many finds have been destroyed in the late 90s. Finally the author thanks the reviewers, especially Prof. Dr. Müller-Beck, for their supporting critics of the first paper draft. And last, the author would like to thank S. Stevens for the spell check.