FRAGMENTATION

Virtually every period in human evolutionary history provides examples of co-evolutionary processes with animal, plant or fungi species. Some of the earliest examples of co-evolutionary processes come from zooarchaeological studies of human interactions with members of the order Carnivora. Archaeological research on this subject goes back 50 years or more and follows numerous conceptual paths. This paper explores ideas and some of the evidence of hominin-carnivore co-evolutionary processes from the viewpoint of evolutionary ecology and the extent to which these ideas have progressed in recent decades. The challenge is to demonstrate that the evolutionary paths of co-evolving species were mutually constrained. Some key behavioral outcomes among humans include non-primate behaviors such as habitual food transport, extensive food sharing and efficient processing of animal foods-behaviors also critical to the success of many of the social carnivores. Another outcome argued to have arisen from co-evolutionary relations is partial complementarity in the patterns of prey age selection among humans and other large predators across African and Eurasian ecosystems. At the heart of ideas about cause in hominin-carnivore co-evolution are competition models.

Previous studies have emphasized the overlap in bone modifications by different types of carnivores. However, the documented overlap is not enough to prevent taphonomists from differentiating among carnivore types (e.g., felids, hyaenids and canids). The present work elaborates on previous experimental works and produces an analogical framework created with the intention of differentiating predator/scavenger bone modification by analyzing furrowing patterns on epiphyseal ends. Taking long bones from the same carcass type as a reference, it will be shown that the three major groups of carnivores (felids, hyaenids and canids) can be successfully differentiated. The patterns of long bone furrowing by these three groups will be presented.

Several models that have been proposed for explaining human evolution involve human-carnivore relationships. Reconstructing the structure and functioning of past food webs is, therefore, essential for evaluating the assumptions and conclusions of these models. Here we present a preliminary attempt to reconstruct the structure of some Pleistocene food webs from the Iberian Peninsula and to compare them with recent food webs from several regions and environments. The present work is a first step towards the reconstruction of past food web dynamics and is aimed at gaining a better understanding of the role of humans in past food webs. Our analysis was restricted to mammals weighing more than 10 kg because they constitute the portion of the food web that allegedly included hominins. Predator-prey interactions for fossil species pairs are inferred from their body sizes, evidence from the fossil record and behavioural information from close living relatives. The number of potential prey per predator in Pleistocene and recent food webs is compared, and the relationship between the number of secondary consumers and the standing biomass of primary consumers, estimated using allometric relationships, is investigated. Pleistocene food webs show a distinctive architecture, with a relatively large number of secondary consumers and a small number of primary consumers. In addition, the size distribution of primary consumers also differs between recent and Pleistocene food webs. Our results point to high intraguild competition during the Pleistocene, especially during the Early Pleistocene, which may have conditioned resource availability for Paleolithic hunter-gatherer populations.

Pleistocene cave sediments present a complex geological, paleontological and archaeological record. Cave strata represent long time-averaged depositional processes that reflect substantial climatic variation. Parsing the occupation of caves by hominids and carnivores during the Middle Palaeolithic is one of the challenges of taphonomic research. Both groups served as accumulators of animal bone in what became palimpsests of repeated occupations. These are low-resolution deposits, making it difficult to discern patterning and spatial organization or the relationship between hominids and carnivores. The Grotte du Bison, Arcy-sur-Cure, France contains a well documented sequence of occupations by Neanderthals and other carnivores within a long geologic sequence that reflects climatic variation. This paper explores the periodicity and frequency of use of the cave by different species over time. Data suggest a low frequency of habitation by hominids and more frequent occupation by hyenas and other large carnivores during the Mousterian, with little inter-species competition for space at any given time. Later in the sequence, during the Châtelperronian, data indicate exclusive occupation by Neanderthals, followed by abandonment of the space to hibernating bears, with a shift in occupation by humans to the neighboring Grotte du Renne.

Large carnivore neotaphonomy is used to provide guidelines for understanding fossil bone assemblages. However, few studies have been carried out on the taphonomic signatures of wolves (Canis lupus) in their natural settings. From 2001 to 2007, 56 wolf feeding places were studied in 2 geographic areas of Poland (Bialowieza, Bieszczady). We recorded ecological aspects such as prey selection, time span of carcasses use, scavengers' activity and the identification of prey from ungulate hairs found in scats, and taphonomic considerations, such as the number and type of bone remains, intensity of tooth modification on carcasses and the effect of digestion on skeletal elements observed in scats. Localities studied included kill sites (4 C. capreolus and 20 C. elaphus in Bialowieza, 29 C. elaphus in Bieszczady) and scavenging sites (10 B. bonasus carcasses in Bialowieza). In order to characterize taphonomically impact of wolf on medium- and large-size ungulates, the general bone modifications recorded in this study are compared with data from North American and Iberian wolf feeding sites as well as from other large carnivore (Crocuta) den contents.

Hominid-carnivore interaction is a constant feature along the Pleistocene: both species shared time and space, and contributed to the formation of bone assemblages. Thus, the identification of the agent responsible for the accumulations found in any site demands a series of analyses. Taking into account that wolves were frequent carnivores in the European Pleistocene as well as potential predators of medium-sized prey, we approach the study of the record they produce on carcasses. Based on previous works of their taphonomic impact on horse carcasses (Yravedra et al., 2011), we now focus on the distribution patterns they generate and the identification of wolves either as dispersal or accumulating agents in order to compare this behaviour with the patterns found at Palaeolithic sites. Our research suggests that wolves are wolves are shown to be agents of dispersal.

Hyaenas are important consumers of meat and accumulators of bone. A number of taphonomic studies have focused on modern and fossil assemblages accumulated by hyaenas with a view to developing greater understanding of palaeontological and archaeological assemblages. This research has revealed important variability in the characteristics of bone accumulations, not only through different methods of study, but also because of field context, intrinsic factors, such as occupation time and age profiles, and extrinsic factors, such as climatic conditions, prey species structure, competition. In order to characterize hyaenas as taphonomic agents, we present a diachronic comparison of the consumption of long bones of medium-to-large ungulates (size categories III to V) by cave hyaenas Crocuta crocuta spelaea based on Middle to Upper Pleistocene assemblages from Lunel-Viel 1 (Marine Isotopic Stage 9-11; NISP=2149), Artenac c10 (MIS 5c; NIPS=136), Peyre (MIS 5e; NISP=330), Fouvent (MIS 3; NISP=194) and Conives (MIS 3; NISP=523). Fragmentation of appendicular elements highlights two features; depending on the element, bones with significant amounts of meat and marrow (humerus, radius, femur, tibia) are widely represented by long shaft fragments with chewing damage, while metapodials, which are consumed to a lesser degree are often complete or have large portions of their shafts intact with often an end affected by scooping out on the caudal side. Comparison of bone accumulations from modern hyaena dens (Djibouti) and bone consumed by other Holocene and Pleistocene predators, Ursus arctos (Mont Ventoux) and Panthera onca gombaszoegensis (Artenac Ens. I and II), highlights the particular taphonomic signature and significant impact of Cave hyaenas on medium-to-large size prey.

Cave bear (Ursus spelaeus) and brown bear (U. arctos) fossils are common in the Eurasian Late Pleistocene deposits. Human presence is often indicated by Mousterian culture artifacts. The cave bear and the Neanderthals (Homo neanderthalensis), a human group associated with Mousterian culture, became extinct before the Holocene, whereas the brown bear survived. Here we studied large mammal paleocommunities from fossil localities with brown bear fossils, cave bear fossils and Mousterian lithic assemblage in the Late Pleistocene to test if paleocommunities reflect different habitats for brown bear than the two extinct species. Second we asked if paleocommunities in sites with Mousterian culture assemblage reflect more the prey selection than the environment of the people. Our results indicate that Mousterian sites have higher abundance of equids and mustelids than the bear sites, but lower abundance of large carnivores, especially cursorial ones. These probably reflect prey preferences and competitive exclusion of large carnivores by people associated with Mousterian lithic culture. We found no significant differences in paleocommunities suggesting different habitats for brown bear, cave bear or for people associated with Mousterian lithic assemblage.

In Eastern Europe, they are many Middle Paleolithic caves, dated to the Last Interglacial of the First Wechselien Interpleniglacial, which delivered both traces of human and ursid occupations. Moreover, this occurs less frequently in hyena dens. By region, three types of archaeological sites have been evidenced: (1) poor in carnivore bones, (2) rich in bones of different carnivorous species, (3) rich in bones of one carnivore species, divided into two types: low anthropogenic occupation (3a) and high anthropogenic occupation (3b). In Eastern Europe, the exploitation of carnivores by Neanderthals is very rare, it appears slightly more intense in layers with industry attributed to the transition and the ancient Aurignacian.

In several caves, lithic artifacts or human-modified bones have been found as more or less associated with large faunal assemblages accumulated by cave hyenas. Even if these small records remain often hard to interpret, they are essential to understand interactions between human groups and other cave dwellers. Their study can bring new elements of discussion on critical issues such as the intensity of competition for shelter occupation or the potential existence of specific human activities in hyena dens (e.g. scavenging of meat scraps, collecting of bones). Here we present an interdisciplinary work on two Upper Pleistocene hyena dens, Camiac and La Chauverie, where a small number of Middle Paleolithic artifacts have been found. Results are provided by the combination of three disciplines: faunal taphonomy, lithic analyses (including studies of reduction sequences) and spatial analysis (threedimensional plotting, systematic refitting). At Camiac and La Chauverie, our interdisciplinary analysis highlights two distinctive types of human occupations. Sites that first seem to be closely related (hyena dens with scarce lithic artifacts) hide in fact a variety of situations, ranging from the succession of independent occupations of human groups and hyenas to potential traces of short human visits to hyena dens. Finally, by comparing our results with the regional record, we discuss the actual evidence for competition for shelter between cave hyenas and the last Neanderthals in southwestern France.

Noisetier Cave (French Pyrenees) has yielded Mousterian artefacts associated with numerous faunal remains. The faunal spectrum is dominated by chamois and ibex followed by red deer and bovids. A previous taphonomic analysis underlined the occurrence of two distinct types of bone accumulations. The red deer, bovid and a part of the ibex remains have been accumulated by Neanderthal. We suspected that the bearded vultures were responsible for the chamois and some of the ibex remains. The study of the carnivore remains illustrated the abundance of teeth and to a lesser extent bones attributed to both young and adult Cuon alpinus individuals. The identification of shed milk teeth demonstrates that this carnivore used the cave as a nursery den. According to several authors dholes never bring back carcasses to their dens in order to protect their offspring from other carnivores. However they tend to select an area inside their den to defecate. We analysed modern scats of wolf in order to constitute a taphonomic referential. Our results strongly suggest that most of the digested remains from the Noisetier Cave come from dhole scats. This carnivore can be considered, as Binford previously suggested, as a bone accumulator and consequently as a new taphonomic agent. Given the numerous sites where the fossil remains of this carnivore were identified we argue that the dholes could have biased the composition of faunal spectrums and maybe our understanding on human subsistence.

Study of faunal series resulting from recent excavations in two caves in North Atlantic Morocco (Grotte à Hominidés - GH - and Grotte des Rhinocéros - GDR - at Thomas I and Oulad Hamida 1 quarries, Casablanca) has yielded new evidence concerning the gathering and processing of ungulates carcasses during the Middle Pleistocene in this part of North Africa. Preliminary taphonomic analysis of the macrofauna indicates that the carcasses were mainly introduced in the caves by carnivores. Additionally, marks generated by porcupines also occur. Dimensions and morphologies of tooth-marks and coprolites suggest that carnivores of different sizes (mainly middle-sized canids, hyenids and felids), as well as porcupines, used the cave. Cut-marks on the bones are absent at GH and scarce at GDR, despite their association with lithic artefacts and human fossils. This raises the question of the relationship between hominins and other competitors in these caves. The recurring question is to determine the modalities of niche partitioning by the various predators and/or carrion-eaters as well as the mode of introduction of artefacts and human remains.

Boxgrove is an exceptional Lower Palaeolithic locality. Fine grained deposits that contain large quantities of lithic tools and modified fauna have been identified and excavated over a large area. These large data sets allow for a unique discussion of hominin-carnivore interactions at a landscape scale. Modifications identified and reported in this study demonstrate that hominins had primary access to most carcasses and products. This primacy can be tracked across the site and relates to animals of different sizes and that inhabited different environmental niches. Where carnivore and hominin modifications have been identified on the same specimen, the former frequently overlie the latter. Furthermore, Boxgrove preserves evidence for the repeated episodes of hominin-carnivore behaviour within and across this landsurface along with evidence for single episode butchery at GTP 17. At Boxgrove there is evidence for direct hominin-carnivore interaction. There is a high intensity and quantity of hominin butchery signatures compared to carnivore modifications. The Boxgrove faunal assemblage clearly indicates that by 500 kya H. heidelbergensis was a top predator in this environment and capable of acquiring and securing prey, of various sizes, against other carnivores such as lion and hyaena.

This paper proposes a novel approach to study the interactions of Neanderthals and carnivores in the cave of Zafarraya by comparing the lithic archaeological and faunal records with a statistical path analysis, taking into consideration the ecology of the main carnivore predators and large herbivore prey foraging in the surroundings of the cave. The results of the analyses confirm and shed further light on previous taphonomic and zooarcheological research. The findings concur with the two-species Lotka- Volterra competition model for resources which stipulates that when niche overlap is complete the species with the larger fitness excludes the other. Our analysis shows that in the immediate vicinity of the cave, the fitness of Panthera was greater than Neanderthals', i.e. when Panthera was present it excluded Neanderthals as evidenced by the record of Capra and Rupicapra remains. It also shows that further in the southern hills and the polje where large herbivores roamed, Neanderthals had a greater fitness than carnivores which translated into their primary accumulation in the cave of remains of Cervus elaphus and other large herbivores. Coexistence from occasional niche overlap is apparent when one or the other predator scavenged, but from a time prospective it must have been short periods linked to seasonality, weather conditions and occupation randomness. In Zafarraya, the archaeological record would indicate that the degree of fitness of the herbivore prey accumulators, carnivores or Neanderthals, was related to the nature of the geomorphological domains in the vicinity of the cave and the favored foraging areas of hunted herbivores.

The cave-site of Furninha is located on the sea cliff face of the small peninsular zone of Peniche in Portuguese Estremadura. Excavated by N. Delgado at the end of nineteenth century, it yielded a very rich and diversified Pleistocene vertebrate assemblage attributed to the early Late Pleistocene. A preliminary global overview is given in terms of geology, palaeobotany, palaeobiology s.l. and taphonomy. The fossils come from a ca.6 m deep pit in the basement of the karstic galleries with several layers. Carnivore remains dominate among large mammals, especially bear, hyena, wolf and lynx. In contrast, herbivore remains are few and show specific damage due to predators. Among the small mammals, a huge number of leporid bones as well as remains of insectivores and birds were found. Some few lithic artifacts occurred. The two first predators, bear and hyena, represent a well preserved collection with several skulls and mandibles as well as many complete long bones from both young and adult individuals. Hyenid remains are attributed to 'Hyaena prisca', a taxon known from late Middle Pleistocene sites, which survived later in Portugal. Its taxonomic and phyletic relationships are not well known, and first morphofunctional analyses precise its status and paleoautoecology. We detail the vertebrate assemblages recovered from the different stratigraphic layers, which will allow us to comment the degree of intra- and interspecific competitions, and to explain the origin of such bone accumulations and behaviors of these predators.

This study offers an overview of carnivore remains from archaeological contexts and provides evidence of interaction between carnivores and hominins in the Swabian Jura during the Middle and Upper Paleolithic (~50,000-27,000 uncal B.P.). First, we present data on the carnivores in the faunal assemblages from the area, followed by a general comparison of anthropogenic and carnivore modifications on faunal remains. Further, we describe some archaeological findings that demonstrate unique ways in which humans interacted with carnivores in the early and middle Upper Paleolithic. This study documents the pattern of carnivore representation in the zooarchaeological assemblages on a regional scale. The analyses of faunal assemblages across five cave sites in the two valleys of the Swabian Jura indicate intensified use of caves by humans, which corresponds to a decreased presence of large and medium sized carnivores at the beginning of the Upper Paleolithic in addition to the use of carnivore figures in the cultural repertoire of the Aurignacian and the increased exploitation of carnivores in the Gravettian period.

Several cuon bones were discovered recently in three Upper Pleistocene archeological sites in the central area of the Iberian Mediterranean. This has proved that there were different types of interactions between dholes and prehistoric human groups. Firstly, evidence found in the archeological sites of Cova Negra and Coves de Santa Maira shows the use of carcasses of dholes by human hunter-gatherers. Secondly, the dhole remains recovered in Cova del Parpalló shows the dholes and humans could occupy the same habitat. In this case, the dhole died by natural causes in a small and isolated gallery before the human groups occupied the cave during the gravettian period. Associated with the dhole bones, there were also many ungulate mammal remains found. Some of these bones shows carnivore tooth marks. Due to these findings, we can presume that the dhole might have been the predator responsible for the bones discovered within the chamber. With the data provided we can come to the conclusion that this species had a more prominent role than we originally thought.

The Búho and the Zarzamora caves (Segovia, Spain) are two small karstic cavities in the North of the Central System Cretaceous limestones, in the transitional region between the Sierra de Guadarrama Mountains and the Castilian Plateau. The infilling sediment was excavated during two periods, from 1988-1990 and from 2008- actuality, and subsequently has been assigned to the Late Pleistocene. The aim of this study is the taphonomical analysis of the macrofaunal remains from the old and the new excavation campaigns. The taxonomical list includes: Carnivora (Crocuta crocuta, cf. Panthera sp., Lynx sp., Canis lupus, Vulpes vulpes and Meles meles), Perissodactyla (Equus ferus, Equus hydruntinus and Stephanorhinus hemitoechus) and Artiodactyla (Sus scrofa, Cervus elaphus, Bison priscus and Bos primigenius). The abundance of hyena juveniles and coprolites, as well as carnivore tooth marks and digested bones suggest that the Búho and Zarzamora caves worked as a spotted hyena den during the Late Pleistocene. Nevertheless some human activity is also present in the Zarzamora cave site with evidences of cut marks in carnivore remains (Lynx sp.). The macro faunal association suggests an open environment where equids were the most abundant herbivores.

Lower IX level from Labeko Koba and X level of Ekain have been considered relevant because their archaeological attribution to the Chatelperronian. Nevertheless the association of these archaeological evidences with complex faunal assemblages, characterized by the high presence of carnivores, requires a detailed archaeozoological analysis in order to understand the real nature of human interaction in the site and thus asses the function of these occupations. The Labeko Koba IX lower layer is an occupation of cave hyenas (Crocuta crocuta spelaea), where we can identify, through a taphonomic analysis that a part of the assemblage had anthropic origin. On the other hand, the level X of Ekain is an accumulation of remains of cave bear (Ursus spelaeus) associated with a small assemblage of lithic artifacts. The particularities of cave bear ethology during hibernation suggest that ursids were not the main accumulator of other species bones. In this paper, we wish to contribute to a better understanding of human presence in these sites during the Chatelperronian, by comparing the results produced by the Archaeozoology and the Lithic Techno-tipology. Grace to this interdisciplinary study it has been possible to identify, in both levels, the role played by carnivores and humans in faunal remain accumulation and thus characterize the impact and nature of human presence in both sites. This paper could be a contribution for understanding the coevolution of humans and carnivores in caves of the southwestern Europe during the early Upper Palaeolithic.

Cave bear (Ursus spelaeus Rosenmüller-Heinroth, 1794) are the most abundant taxon in the lower levels of many archaeological sites in Cantabrian Area. Through the scientific literature, archaeological levels have been consistently assigned to the different cultural periods, depending on the identified stone tools. In this paper, we would like to contribute to the interpretation of these sequences, through the accurate analysis of their archaeozoological accumulations. By presenting three examples, Lezetxiki, Lezetxiki II and Astigarragako Kobea, we will try to bring new data to this problem. Archaeozoological analysis carried out at these three stratigraphical sequences, have shown the existence of intense bears (Ursus spelaeus deningeroide Mottle, 1964 and Ursus spelaeus Rosenmüller-Heinroth, 1794) occupation and human groups ephemeral presence (through the presence of lithic implements), in the oldest levels (Lower Palaeolithic) stratigraphic series of the three cavities. Our aim is to present the preliminary archeozoological and taphonomic results of these three sequences.

This work describes the fossil accumulations recovered during the excavation of the sinkhole called Sima de Abraham (Sierra Alcaide, Priego de Córdoba, Iberia). This site is characterized by a prominent accumulation of mammalian fossil remains, including carnivores (especially bears, lynx, wildcats and spotted hyenas) and artiodactyls (red deer and ibex) among other species. Human activity does not seem to be the primary agent of accumulation, although a series of cut marks have been found on the articular cavity of the proximal ulna of a large bear, providing interesting insight into the interaction between bears and humans. The age of the deposit was set in the Late Pleistocene in a previous study according to the presence of the southern water vole (Arvicola sapidus) and subsequently corroborated by means of AMS and U/Th, obtaining an approximate age of 40- 30 ky BP, corresponding to MIS 3, in the latest Pleistocene.

Fossil evidence about the interaction between carnivores and the first human colonizers of southern South America is presented. The time overlap of carnivores and humans in this region is discussed using the available Late Pleistocene radiocarbon chronologies. On the other hand, the selection of places to live in the regional space is evaluated. Cases in which both carnivores (Panthera onca mesembrina, Smilodon sp., Arctotherium tarijense) and humans are present in the same sites-deposits are then considered and it appears that their interaction was not important. Evidence for human utilization of sites dominated by carnivores is ephemeral at most, while the presence of carnivores at sites dominated by humans is never important. The faunal record from both classes of sites indicates that carnivores as well as humans shared some subsistence resources (Hippidion saldiasi and camelids). However, they differ in the selection of living spaces. Carnivores selected endogenous caves, while humans tend to use exogenous caves.