Quaternary Science Reviews 308 (2023) 108031 Contents lists available at ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev To den or not to den. Contributions to the taphonomic history of the Early Pleistocene site of Venta Micena 4 (Orce, Guadix-Baza Basin) Jose Yravedra Sainz de los Terreros a, b, c, d, e, *, Carmen Luzo  n f, Jose A. Solano f, Gonzalo J. Linares-Matas g, h, Vero  nica Estaca-Gomez a, Juan Jose  Rodríguez-Alba a, Lloyd A. Courtenay , Darío Herranz-Rodrigo , Alexia Serrano-Ramos e, Jose a, i b  Ca mara e, j, j Juha Saarinen , Christian Sa nchez-Bandera , Hugues-Alexandre Blain , k , l k , l Suvi Viranta-Kovanen , Daniel DeMiguel n, o, Beatriz Azanza p, Oriol Oms q, m nez Arenas e, f, u, v, ** Jordi Agustí k, l, r, Mikael Fortelius j, s, t, Juan Manuel Jime a Departamento de Prehistoria, Historia Antigua y Arqueología, Universidad Complutense de Madrid, Madrid, Spain b C.A.I. Unidad de Arqueometría y Ana lisis Arqueologico, Universidad Complutense de Madrid, Madrid, Spain c Grupo de Investigacion Ecosistemas Cuaternarios, Universidad Complutense de Madrid, Madrid, Spain d Grupo de Investigacio n Arqueología Prehisto rica, Universidad Complutense de Madrid, Madrid, Spain e Museo Primeros Pobladores de Europa ‘Josep Gibert’, Orce, Granada, Spain f Department of Prehistory and Archaeology, University of Granada, Granada, Spain g Emmanuel College, University of Cambridge, UK h School of Archaeology, St. Hugh's College, University of Oxford, UK i Department of Cartographic and Land Engineering, Higher Polytechnic School of Avila, University of Salamanca, Avila, Spain j Department of Geosciences and Geography, University of Helsinki, Helsinki, Finland k  de Palaeoecologia Humana i Evolucio Institut Catala  Social (IPHES-CERCA), Tarragona, Spain l Universitat Rovira i Virgili, Departament D’Histo ria i Histo ria de L’Art, Tarragona, Spain m Department of Anatomy, University of Helsinki, Finland n ARAID/Departamento Ciencias de La Tierra, Universidad de Zaragoza, Zaragoza, Spain o  de Paleontologia Miquel Crusafont, Universitat Auto Institut Catala noma de Barcelona, Barcelona, Spain p Departamento de Ciencias de la Tierra, Universidad de Zaragoza, Instituto Universitario de Investigacio n en Ciencias Ambientales de Arago n (IUCA), Zaragoza, Spain q Departament de Geología, Universitat Auto noma de Barcelona, Cerdanyola Del Vall es, Spain r ICREA, Barcelona, Spain s Natural Sciences Unit, Finnish Museum of Natural History - LUOMUS, Helsinki, Finland t Departamento de Prehistoria and Arqueología, Universidad de Granada, Granada, Spain Visiting Scholars Program u Instituto Universitario de La Paz y Los Conflictos, Universidad de Granada, Granada, Spain v Gibraltar National Museum, Gibraltar, United Kingdom a r t i c l e i n f o a b s t r a c t Venta Micena (Orce, Guadix-Baza Basin, Spain) is an Early Pleistocene locality renowned for the richness Article history: and quality of its palaeontological record. VM is spread over an area of 2.5 km2, where several exposed Received 8 June 2022 fossil outcrops are visible amidst its gorges and ravines. The best known of these sites, VM3, has been Received in revised form 27 January 2023 interpreted as a hyaena den. In addition, a new site, named VM4, has recently been the focus of fieldwork Accepted 28 February 2023 and taphonomic studies. The publication by Luzo  n et al. (2021) pointed out that VM4 presents a more Available online xxx complex history than VM3. First, two different sub-levels were identified: VM4-I and VM4-II. Secondly, the preliminary taphonomic analysis showed conspicuous differences with regard to VM3. Nevertheless, Handling Editor: Danielle Schreve such interpretation has been challenged by Palmqvist et al. (2022) who proposed that VM3 and VM4 are both the result of a single depositional process, entailing the selective transport of skeletal parts by the giant extinct hyaena Pachycrocuta brevirostris back to its den. Using well-preserved faunal elements Keywords: whose depositional context and provenance are reliable, in this paper we show that: 1) there are two Herbivores Carnivores clearly defined sub-levels in VM4 with some shared taphonomic characteristics as well as some notable Modifications differences; 2) VM3 and VM4 exhibit enough divergence to support differences in site formation * Corresponding author. Departamento de Prehistoria, Historia Antigua y Arqueología, Universidad Complutense de Madrid, Madrid, Spain. ** Corresponding author. Departamento de Prehistoria y Arqueología, Universidad de Granada, Granada, Spain. nez Arenas). E-mail addresses: jyravedr@ucm.es (J. Yravedra Sainz de los Terreros), jumajia@ugr.es (J.M. Jime https://doi.org/10.1016/j.quascirev.2023.108031 0277-3791/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Depositional events processes; 3) The interpretation of both VM4-I and VM4-II is more consistent with their characterisation Open-air sites as open-air sites in which multiple agents and depositional processes contributed to its formation, rather than with hyaena dens. Nevertheless, excavations are still in progress at VM4 and therefore any results and interpretations ought to be considered as provisional. © 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction Palmqvist et al. (2022) show several inconsistencies and inaccura- cies that need to be mentioned. Among all the sectors discovered at Venta Micena (VM) (Orce, Guadix-Baza Basin) is an important VM, Corte III ¼ VM3 (excavated between 1982 and 1995) stands out. Early Pleistocene palaeontological complex in southeastern Spain, Different authors have reported different values for the total well-known for its extensive and taxonomically diverse macro- excavated surface: Espigares (2010) noted that 326 m2 were exca- vertebrate fossil assemblage (e.g. Martínez-Navarro, 1991; Luzo n vated, whereas Palmqvist et al. (2022) gave a reduced value of et al., 2021; Palmqvist et al., 2022). The discovery of a cranial 220 m2. Similarly, the size of the fossil sample retrieved differs fragment (VM-0) initially attributed to the genus Homo (Gibert slightly depending on the author (Table 1). Regardless of the actual et al., 1983) put VM in the international spotlight. Beyond debates number of remains, the material retrieved from this fossil assem- spurred in relation to this controversial specimen (e.g. Gibert et al., blage has formed the basis of dozens of scientific publications, 1983; Gibert and Palmqvist, 1995; Borja et al., 1997; Moya -Sola  and turning VM into a classic site for understanding Quaternary Ko€ lher, 1997; Palmqvist, 1997; Martínez-Navarro, 2002; Carandell Palaeontology in Europe and Early Pleistocene carnivore palae- Baruzzi, 2020), VM has become a flagship case-study in Pleisto- oecology, particularly regarding the taphonomic characterisation of cene taphonomy as one of the main sites for studying the behaviour P. brevirostris (Arribas and Palmqvist, 1998, 1999; Palmqvist and of Pachycrocuta brevirostris, alongside Fonelas (Arribas and Garrido, Arribas, 2001; Espigares, 2010; Palmqvist et al., 2011, 2022). 2008; Garrido et al., 2010), Vallparadis (Madurell-Malapeira et al., From a taphonomic perspective, the large amount of bone re- 2011; 2017), Cueva Victoria (Madurell-Malapeira et al., 2015), mains recovered from this sector (VM3) has allowed researchers to Tsiotra Vryssi (Katsagoni et al., 2022) and others (Iannucci et al., conclude that this locality represents a hyaena den or Pachycrocuta 2021). This extinct giant hyaena has often been characterised as a camp, to where adult Pachycrocuta individuals selectively trans- scavenger that specialised in the exploitation of carcasses left by ported and intensively altered many skeletal remains of large her- other carnivores, selectively transporting skeletal parts back to bivores. Most of these carcasses may have been hunted by other their dens (e.g. Arribas and Palmqvist, 1998, 1999; Palmqvist and carnivores, particularly large sabre-toothed felids and canids, based Arribas, 2001; Espigares, 2010; Palmqvist et al., 2011, 2022; on the mortality profiles and skeletal part profiles reported by Perez-Claros and Coca-Ortega, 2020). Nonetheless, other authors Palmqvist et al. (1996), and Arribas and Palmqvist (1998, 1999). The contest this interpretation of Pachycrocuta as a specialised scav- role of VM3 as a hyaena den is further reinforced by the presence of enger, proposing instead that this species could have exhibited numerous coprolites attributed to P. brevirostris (Palmqvist et al., some hunting behaviour as part of a wider opportunistic strategy 1996; Arribas and Palmqvist, 1998) as well as remains from juve- including the ability to kill their own prey and acquire carcasses nile and immature hyaena individuals (Palmqvist et al., 1996, 2011, from other carnivores (Vinuesa et al., 2016; Iannucci et al., 2021). 2022; Arribas and Palmqvist, 1998; Palmqvist and Arribas, 2001). This interpretation finds support in the works by Turner and Anto n Lastly, a recent paper has characterized the morphology of the (1996) and Dennell et al. (2008), who suggested that the greater tooth marks of a small sample from VM3, revealing that the main corpulence, size and body mass of Pachycrocuta brevirostris, modifying agent is the large Pleistocene hyena P. brevirostris alongside a gregarious social behaviour, may have favoured this (Yravedra et al., 2022b). species in their hunting capabilities, although the extent to which Pachycrocuta was a social carnivore and its relationship to their food procurement strategies is still subject of debate among palae- 1.1. Venta Micena 4 (VM4) obiologists and palaeoneurologists (Holekamp et al., 2007a, 2007b, 2015; Arsznov et al., 2010; Sakai et al., 2011; Vinuesa et al., 2015, Even though Sondeo IV of Venta Micena was discovered by J. 2016). Gibert and it was known since the 1980s (Agustí et al., 1985), it is The history of research at Orce goes back to the 1970s. The only recently that it has been recognised as an independent site, discovery of VM took place in 1976 by three members of the being renamed as VM4 (Granados et al., 2021; Luzo  n et al., 2021). Instituto Provincial de Paleontología (Sabadell, Barcelona, Spain): VM4 is located 350 m away from VM3 (Fig. 1). To this date, 39 m2 Josep Gibert, Jordi Agustí and Narcís S anchez (Carandell Baruzzi, have been excavated over the course of seven field seasons: 2005, 2013). However, systematic fieldwork was not carried out at the 2013e2015, and 2017e2019 (Luzo  n et al., 2021). site until 1982. Palaeontological interventions were conducted at The stratigraphic sequence of VM is similar in all sectors. It has several areas of VM (Cortes IeV and Sondeos IeV; VM4 ¼ Sondeo been defined on the basis of different sections and it consist of six IV), which provided many macrovertebrate fossils (Agustí et al., units (A-F) (Anadon et al., 1986, 1987). Among these, Unit C, with a 1985; Espigares, 2010). The Early Pleistocene sites of Barranco thickness between 2.3 and 3.4 m, has yielded the largest quantity Leon (BL) and Fuente Nueva 3 (FN3) were subsequently discovered and quality of mammal fossils. This unit is divided into three in- in the wider landscape (e.g. Tixier et al., 1995; Turq et al., 1996), and tervals C0eC1eC2 (Anado n et al., 1986, 1987) (Fig. 2). C0 is char- they both have yielded unquestionable evidence of human pres- acterised by a whitish sandy lutitic limestone while units C1 and C2 ence and activities (e.g. Gibert et al., 1998, 1999; Barsky et al., 2010; consist of around 1.5 m of micritic limestone of palustrine and Toro-Moyano et al., 2013; Ribot et al., 2015; Titton et al., 2018, 2020, lacustrine origin (known as Venta Micena limestone or VM Stra- 2021; Yravedra et al., 2021, 2022a). tum, which surges intermittently over 2.5 km2). The two richest Previous studies carried out on VM3 and synthesized in bone assemblages (VM3 and VM4) are located along Unit 2-Interval C1 limestone, which presents a thickness between 80 and 120 cm 2 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 1 Number of identifiable and unidentifiable fossil remains from VM [VM3 according to Palmqvist et al. (2022)]. Number of identifiable specimens Number of unidentifiable Reference specimens 6454 from Table 2 (6453 from text): 5822 anatomically and taxonomically identified þ 632 anatomically >10,000 Arribas and Palmqvist identified (1998) 6884: 5960 anatomically and taxonomically identified þ >10,000 Espigares (2010) 924 anatomically identified 8150: 6331 anatomically and taxonomically identified þ 1819 anatomically identified “Several thousands” [of shaft Palmqvist et al. (2022) bones] (Fig. 1). Granados and collaborators have shown slight strati- (Agustí et al., 2007, 2010), up to c.1.6 Ma. graphical and geochemical differences between the upper intervals Even though Espigares's Ph.D. dissertation (2010) doctoral at Unit 2 (C1 and C2), as well as between the two palaeosurfaces thesis includes an analysis of the orientations and inclinations of VM4-I and VM4-II (Granados et al., 2021). This notion is reinforced fossil remains from Sondeo IV of VM (n ¼ 211), Luzo n et al. (2021) by the definition of two stratigraphic and sedimentological were the first to offer a comprehensive taphonomic study of boundaries corresponding to two different depositional episodes in VM4. Their research includes an analysis of the remains derived VM4 Unit 2-C1 (Luzo  n et al., 2021). from the 2005, 2018 and 2019 field seasons (NISP ¼ 1609). Despite VM4 would have been located near the edge of a lacustrine the spatial proximity, stratigraphic correlation (Granados et al., environment centered around the relatively large and saline 2021), and the existence of taxonomic parallels between VM3 endorheic Baza lake (Oms et al., 2011; Granados et al., 2021). Unit C and VM4 (Luzo  n et al., 2021; Palmqvist et al., 2022), these two sites would correspond to a regressive phase of the lake, which fostered show some notable differences, according to Luzo  n et al. (2021). surgencies of subterranean freshwater springs, thus favouring the development of an attractive ecosystem for the faunal species 1. The VM4 faunal assemblage exhibits a lower degree of bone found at VM4 and other nearby sites (Luzo  n et al., 2021; Granados weathering than VM3, which suggests a more reduced envi- et al., 2021). The Venta Micena Unit C faunal assemblage comprises ronmental exposure timespan. a series of typical Late Villafranchian macro- and micromammal 2. The VM4 faunal assemblage shows a lower intensity of carnivore species (see Luzo  n et al., 2021; Palmqvist et al., 2022 for different damage. taxonomic proposals). 3. VM4 should be interpreted as a fossil assemblage accumulated From a chronological point of view, Unit C of VM has been dated in the vicinity of a freshwater source where herbivores and using the combined uranium serieseelectron spin resonance carnivores interacted, albeit it is not possible to determine at method, and the mean age is 1.37 ± 0.25 Ma (Duval et al., 2012). this stage which carnivores intervened in the formation and Nevertheless, such dating has been refined and extended through modification of the assemblage, and to what extent. biochronology, on the basis of the presence of Allophaiomys ruffoi Fig. 1. Left: Location of Venta Micena VM (Guadix Baza basin, Orce, Spain). Right: geospatial and stratigraphic relationship between the VM3 and VM4 palaeontological localities. A: Shows the spatial ubication of VM3 and VM4. B: Shows the stratigraphic correlation between VM3 and VM4, C: Satellite image (source: Google EARTH and Instituto Geogra fico Nacional, Spain) indicating the location of the different Venta Micena palaeontological localities (VM1-VM4) (Figure modified from Granados et al., 2021). 3 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Fig. 2. Feeding and furrowing sequences on long limb bones at VM4. 2a: Feeding sequence on equid tibiae. 2b-e distal humerus epiphyses of different species, b: artiodactyl, c: Stephanorhinus, d: Equid, e: Metacervoceros. Scarce furrowing is documented on the distal epiphysis, since it was only observed on the Stephanorhinus' latero-caudal condyle. 4 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 4. VM4 would not constitute a hyaena den or Pachycrocuta camp, posterior portions (carpals, tarsals, and metapodials). on the basis of an absence of immature remains of P. brevirostris, The nature and extent of bone fragmentation was assessed the low frequencies of bones with tooth mark alterations, and based on three variables. First, bones were divided into several size the presence of multiple bones still in anatomical connection. categories based on their length:  3 cm, 3.1e5.0 cm, 5.1e10 cm and 5. VM4 presents two depositional events, whereas VM3 has long 10 cm. Second, long bones were classified according to whether been interpreted as being constituted by a single depositional they present green or dry fractures, following the criteria outlined process. in Villa and Mahieu (1991). Third, the percentage of preserved shaft circumference was defined as follows: Type 1 are specimens that This proposal has recently been disputed by Palmqvist et al. preserve <50% of the shaft circumference intact; Type 2 are spec- (2022). Reinterpreting the results generated on the basis of the imens with >50% of the shaft circumference; and Type 3 are materials retrieved during the 2005 and 2013e2015 field seasons specimens with >75% of shaft circumference, following Bunn (NISP ¼ 3961) and comparing them with the data from VM3, these (1982). authors claim that both sites have different sedimentological bi- Bone alteration patterns were quantified using NISP values, and ographies but a shared functional nature as a Pachycrocuta den. the observation of bone surface modifications was carried out using With the aim of addressing the debate surrounding the nature of both binocular and hand-held lenses. The nature and intensity of VM4, we present in this paper the first comprehensive study of the hydric activity and other waterborne alterations was characterised totality of the VM4 faunal assemblage, including the remains basing on the frequencies of bones with abrasion, polishing and retrieved from the 2005, 2013e2105 and 2017e2019 field seasons. rounding. Abrasion is produced in situations where water may have Since the global taphonomic context of the VM4 assemblage has transported small sedimentary particles which erode bone sur- already been characterised by Luzon et al. (2021), this paper focuses faces, according to Lyman (1994) and Thompson et al. (2011), hydric on more specific aspects related to the biological processes that led polishing and rounding are produced in contexts where bones to the accumulation of faunal remains in the two VM4 sublevels. experience movement or transportation by water flows (Shipman, 1981; Lyman, 1994; Ferna ndez-Jalvo and Andrews, 2016), although 2. Materials and methods sometimes the action of water-suspended particles can also result in the erosion of bone surfaces, producing rounding without the In this paper, the sample has been analysed from a taphonomic bone having experienced displacement (Thompson et al., 2011). perspective. The bone sample analysed at VM4 consists of 4898 These waterborne alterations were each subdivided into three remains from two levels, VM4-I with 1732 remains (Table 2) and different stages: light (the bone surface very little; the bones do not VM4-II with 3166 remains (Table 3), spread over a surface area of have rounded edges, the bone surface can be seen very well, and 39 m2 (see Luzo  n et al., 2021). Micromammal remains were polish is only reflected in a slight shine on the bone), intermediate excluded from this study. The 181 bones that were not assigned to (when rolling edges are already appreciated, the polishing is either level VM4-I or VM4-II have not been included in the study, greater, with shine across the entire bone surface, but bone surface and neither have been included those currently undergoing resto- modifications can still be observed, although with difficulty), and ration work. intense (when the bones are very rolled and eroded, which pre- The taxonomical profiles have been quantified on the basis of vents the documentation of bone surface modifications). NISP (Number of Identified Specimens) and MNI (Minimum Num- Bone weathering was classified following Behrensmeyer (1978). ber of Individuals). MNI estimates have taken into consideration Other processes such as manganese staining, oxidation and element laterality as well as their ontogenetic age (Brain, 1969), biochemical alterations were also quantified as presence/absence. epiphyseal fusion, long bone biometrics and, if applicable, dental Finally, carnivore activity was analysed through the frequency and wear. When the taxonomic attribution of individual bones could anatomical distribution of tooth marks, the pit/score ratio and the not be discerned but they were identifiable anatomically, these quantity of tooth marks tooth marks per bone. Tooth marks were fossils were classified to weight/size classes using different cate- divided into pits, scores, and punctures, and their identification was gories. These follow the scheme provided in Luzo n et al. (2021), made following Binford (1981) and Blumenschine (1995). Accord- where herbivores were assigned to 5 different size classes: Very ing to Capaldo and Blumenschine (1994) and Mocla n et al. (2019), Small Size (0) for species less than 25 kg of weight; Small Size (1), notches were classified into the following types (see Suppl. Data, including species weighing 25e50 kg; Small-Intermediate Size (2), Fig. 1): Type 1 (complete single notches); Type 2 (double opposing including species weighing 50e125 kg; Intermediate Size (3), complete notches); Type 3a (incomplete notches where inflection including species weighing 125e500 kg, divided in 3a points are missing); Type 3b (complete and other incomplete (125e250 kg) and 3b (250e500 kg); Large Size (4), including spe- notches where one inflection points is missing); Type 4 (double cies weighing 500e1000 kg; and Very Large Size (5), species overlapping notches); Type 5 (notches with other double opposing weighing >1000 kg. Carnivora were classified according to three notches); Type 6 (pseudo notches); and Type 7 (notches smaller groups; Small Carnivora (e.g. fox); Intermediate Carnivora (e.g. than 1 cm). Moreover, we have also analysed the angles of green wolf); and Large Carnivora (e.g. lion). The mortality profiles were bone fractures, following Alca ntara García et al. (2006), for they can established by assigning suitable remains to one of four different offer insights into the nature of the force that led to the fracture, ages classes (infantile, juvenile, prime adult-adult and senile), namely dynamic or static load; the latter is typical of carnivore based on tooth eruption, crown wear and epiphyseal fusion. bone breaking activities. To discriminate between the agency of Skeletal profiles were generated using the MNE (Minimum different carnivores, particularly between hyaenas and canids, we Number of Elements) following Marean and Kim (1998) for shaft have followed Mocla n et al. (2019). Lastly, we have also measured classification. We have divided the skeleton into different furrowing intensity following the taphotype typology of anatomical sections, according to the scheme proposed by Yravedra Domínguez-Rodrigo et al. (2015). and Domínguez-Rodrigo (2009), namely: cranial (horns, crania, In order to evaluate the dependence of a given variable (per- maxillae, mandibles and teeth), axial (vertebrae, ribs, pelvises and centage of immature individuals) respect to an independent vari- scapulae); appendicular limbs were subdivided into upper ele- able (adult body mass), we have carried out regression analyses by ments (humerii, femora), intermediate appendicular (radii, ulnae, the adjustment technique of Ordinary Least Square (OLS). If the null tibiae, patellae, fibulae), lower appendicular elements, and hypothesis [H0 (slope ¼ 0)] cannot be rejected (that is, p-value 5 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 2 Taxonomical profiles (VM4-I). S: Senile; A: Adults; J: Juvenile; I: Infants. VM4-I NISP % Herbivores % total MNI % Herbivores % total S/A/J/I % non-adults Mammuthus meridionalis 3 1.01 0.17 1 3.57 2.50 0/0/0/1 100 Stephanorhinus etruscus 13 4.36 0.75 4 14.29 10.00 0/1/1/2 75 Equus altidens 126 42.28 7.27 7 25.00 17.50 1/4/1/1 28.7 Hippopotamus antiquus 17 5.70 0.98 2 7.14 5.00 0/1/0/1 50 Bison sp. 20 6.71 1.15 2 7.14 5.00 1/1/0/0 0 Soergelia minor 8 2.68 0.46 1 3.57 2.50 0/1/0/0 0 Capra alba 16 5.37 0.92 3 10.71 7.50 1/2/0/0 0 Praemegaceros verticornis 39 13.09 2.25 4 14.29 10.00 1/1/1/1 50 Metacervoceros rhenanus 33 11.07 1.91 4 14.29 10.00 1/1/1/0 33.3 Bovidae 23 7.72 1.33 Total Herbivore determinable 298 17.21 Herbivore indet. size 0 5 0.29 Herbivore indet. size 1 1 0.06 Herbivore indet. size 2 60 3.46 Herbivore indet. size 3 97 5.60 Herbivore indet. size 3a 42 2.42 Herbivore indet. size 3b 151 8.72 Herbivore indet. 956 55.20 Total Herbivore 1610 92.96 28 100 70.00 NISP % Carnivora % total MNI % Carnivora % total S/A/J/I % non-adults Canis mosbachensis 7 9.72 0.40 2 22.22 5.00 0/2/0/0 0 Xenocyon (¼ Lycaon) lycaonoides 13 18.06 0.75 1 11.11 2.50 0/1/0/0 0 Vulpes alopecoides 2 2.78 0.12 1 11.11 2.50 0/1/0/0 0 Pachycrocuta brevirostris 16 22.22 0.92 2 22.22 5.00 1/1/0/0 0 Homotherium latidens 1 1.39 0.06 1 11.11 2.50 0/1/0/0 0 Ursus etruscus 4 5.56 0.23 2 22.22 5.00 1/1/0/0 0 Canidae 13 18.06 0.75 Felidae 4 5.56 0.23 Carnivora 12 16.67 0.69 Total Carnivora 72 4.16 9 100.00 22.50 NISP % other % total MNI % other % total S/A/J/I % not adults Lagomorpha 7 14.00 0.40 1 2.50 2.50 0/1/0/0 0 Testudines 38 76.00 2.19 1 2.50 2.50 0/1/0/0 0 Aves 5 10.00 0.29 1 2.50 2.50 0/1/0/0 0 Total 1732 100.00 100.00 40 100 100.00 >0.05), we will consider that there is no significant relationship M. meridionalis, S. etruscus, and H. antiquus are the scarcest group. between the variables used. Regression functions were estimated Regarding carnivores, canids are the most representative group using the statistical package SPSS 28.0. (50% of the carnivore MNI sample), followed by hyaenids, ursids, and felids, all of which exhibit similar frequencies. 3. Results The mortality profiles of both palaeosurfaces (VM4-I and VM4- II) show a predominance of adult individuals in all species except 3.1. Taxonomical and mortality profiles for Size 5 animals. Additionally, we have noticed some differences on which taxa are better represented by immature individuals: VM4-I (Table 2) comprises a faunal assemblage of 1732 remains, rhinoceroses, elephants and hippopotamuses for VM4-I (Table 2) with herbivores representing 70% and carnivores 22.5% of the MNI and elephants for VM4-II (Table 3). Among size 2e3 herbivores, (Table 2). Among the herbivore taxa, the equid E. altidens is the only P. verticornis (VM4-I: 50% and VM4-II: 55.5%) and M. rhenanus predominant species (25% of individuals), but cervids are the most (VM4-II: 50%) exhibit an important representation of immature abundant group (28.6%), represented by genera Praemegaceros and individuals, while adults dominate the mortality profiles of all Metacervoceros (Table 2). Rhinocerotids and caprids are in third other taxa. Thus, we can establish that at VM-4 the largest species place (each accounting for 14% of the MNI), followed by large bo- are dominated by infant-juvenile individuals, while small and vids, such as bison, hippopotamids with 7%, and lastly pro- medium species are dominated by adults. Concerning bone pa- boscideans with 3.5%. Among carnivores, C. mosbachensis, P. thologies, only a horse metatarsal and a bear phalanx have been brevirostris and U. etruscus are represented by two individuals each, identified with signs of osteoarthritis, as well as some cranial while the remaining carnivores are only represented by a single fragments (Suppl. Data, Fig. 2). individual. VM4-II has a faunal assemblage of 3166 remains, with herbi- 3.2. Skeletal profiles vores representing 69% of the MNI (Table 3). The most represented species are E. altidens and P. verticonis, each with 22.5% of the MNI. Size 3 species show the most representative skeletal profiles at Cervids (both Praemegaceros and Metacervoceros) are also the pre- both VM4-I (Table 4) and VM4-II (Table 5), while other species are dominant group with 37.5% of the total herbivore assemblage and less abundant. Size 2 taxa present lower frequencies than Size 3 26% of the total MNI. Bovids comprise the third most numerous species, though bones from every anatomical portion have been group, followed by the caprids C. alba and S. minor, and mega- recorded. However, very few remains can be attributed to Sizes 1 herbivores and 5. There is a balanced representation of the different anatomical 6 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 3 Taxonomical profiles (VM4-II). S: Senile; A: Adult; J: Juvenile; I: Infant. VM4-II NISP % Herbivores % total MNI % Herbivores % total S/A/J/I % non-adults M. meridionalis 6 1.16 0.19 3 7.50 5.36 0/1/1/1 66.6 S. etruscus 14 2.70 0.44 2 5.00 3.57 0/2/0/0 0 E. altidens 177 34.17 5.59 9 22.50 16.07 1/5/2/1 33.3 H. antiquus 4 0.77 0.13 1 2.50 1.79 0/1/0/0 0 Bison sp. 70 13.51 2.21 5 12.50 8.93 0/3/1/1 40 H. cf. gracilis 7 1.35 0.22 1 2.50 1.79 0/1/0/0 0 S. minor 18 3.47 0.57 1 2.50 1.79 0/1/0/0 0 C. alba 23 4.44 0.73 3 7.50 5.36 1/1/1/0 33.3 P. verticornis 89 17.18 2.81 9 22.50 16.07 1/3/1/4 55.5 M. rhenanus 73 14.09 2.31 6 15.00 10.71 1/2/2/1 50 Cervidae 2 0.39 0.06 Bovidae 35 6.76 1.11 Total Herbivore determinable 518 16.36 Herbivore indet. Size 0 8 0.25 Herbivore indet. Size 1 2 0.06 Herbivore indet. Size 2 74 2.34 Herbivore indet. Size 3 91 2.87 Herbivore indet. Size 3a 9 0.28 Herbivore indet. Size 3b 112 3.54 Herbivore indet. 2095 66.17 Total Herbivore 2909 100.00 93.39 40 100 71.43 NISP % Carnivora % total MNI % Carnivora % total S/A/J/I % non-adults C. mosbachensis 26 11.98 0.82 3 25.00 5.36 1/2/0/0 0 X. (¼ Lycaon) lycaonoides 71 32.72 2.24 3 25.00 5.36 1/2/0/0 0 P. brevirostris 7 3.23 0.22 2 16.67 3.57 1/1/0/0 0 H. latidens 2 0.92 0.06 1 8.33 1.79 0/1/0/0 0 M. cultridens 3 1.38 0.09 1 8.33 1.79 0/1/0/0 0 U. etruscus 11 5.07 0.35 2 16.67 3.57 1/1/0/0 0 Canidae 77 35.48 2.43 Felidae 2 0.92 0.06 Carnivora 18 8.29 0.57 Total Carnivora 217 100.00 6.85 12 100.00 21.43 NISP % other % total MNI % other % total S/A/J/I % non-adults Lagomorpha 21 44.68 0.66 2 3.57 3.57 0/2/0/0 0 Testudines 3 6.38 0.09 1 1.79 1.79 0/1/0/0 0 Aves 16 34.04 0.51 1 1.79 1.79 0/1/0/0 0 Total 3166 100 100.00 56 100 100.00 portions among Size 3 species at VM4-I: axial remains are repre- ratios exhibited by limb bone epiphyses, since they are far removed sented by 32 elements, upper and intermediate limb bones have 31 from the value of 1 that would indicate a complete representation. elements each, and there are 37 lower appendicular elements When measuring the relative abundance of the less dense epiph- (Table 4). The only notable difference is documented between yses (proximal humerus plus distal radius)/(distal humerus plus forelimb and hindlimb bones, with the latter being more abundant proximal radius) we obtain values of 0.125 for VM4-I and 0.24 for among Size 3 animals; this trend is reversed for Size 2 animals, with VM4-II, evidencing the skeletal bias of the sample. Similar levels of forelimbs being more commonly depicted (Table 4). skeletal bias are documented based on the predominance of dia- At VM4-II, the faunal sample is larger and more representative, physiary elements over epiphyses (Supp. Data, SF6). Despite these even though it is still dominated by teeth (Table 5). Species from biases, the representation of all skeletal parts among Size 2 and Size Sizes 2e3 are the most abundant, with all skeletal portions repre- 3 animals may indicate that some animals were perhaps initially sented, although there are some notable biases. For example, upper complete at the site. limb bones are poorly represented (N ¼ 4) while other portions reach MNE values of 11 (Table 5). Both forelimbs and hindlimbs are represented, although the former are slightly more common. 3.3. Taphonomic evidence Metacarpals are particularly well represented among Size 3 ani- mals. For species from Sizes 4e5, the sample is limited, and only Bone surface preservation is generally good both at VM4-I (76%) forelimbs are present. and VM4-II (84%) (Table 6). Weathering alterations are low at both When comparing VM4-I and VM4-II, the skeletal part profiles levels, which implies that bones were not subjected to subaerial are biased for all species, particularly when comparing MNEs exposure. The frequencies of manganese or other oxide staining, as (Tables 4 and 5) with the MNI values for Sizes 2, 3 and 5 (Tables 2 well as bone concretions, are very low (Table 6). Lastly, in relation to and 3). In this regard, Size 3 animals at VM4-I would be represented waterborne alterations, rounding and polishing have not been on average by 17.3 elements per individual, Size 2 species by 8.6 observed, while abrasion is documented on 29% of the bone elements per individual, and Size 5 by 1.6 elements per individual. assemblage from VM4-I and 22% at VM4-II. However, intense At VM4-II, these ratios are still rather biased, with 16.1 elements per abrasion is only documented on 8,6% and 5,9% of the sample individual for Size 3 species, 17.4 elements per individual for Size 2 (Table 6). Moreover, biochemical alterations affect 17.6% and 11.2% animals, and 3.3 elements per individual for Size 5 animals. of VM4-I and VM4-II, respectively (Table 6). This bias is further documented in the low representativity VM4-I and VM4-II present a moderate degree of fragmentation, with over 50% of remains being larger than 3 cm (Table 6). This 7 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 4 Skeletal part profiles at VM4-I in relation to the MNE. MNE VM4-I 0 1 2 % 3 % 3a % 3b % Total 3 % 4 5 Carniv 1 Carniv 2 Carniv 3 Indet. Horn-antler 1 1.45 0.00 0.00 0.00 0 0.00 Cranial 0.00 1 2.94 0.00 0.00 1 0.53 7 Maxillar 0.00 0.00 1 3.85 0.00 1 0.53 Mandible 5 7.25 1 2.94 1 3.85 2 1.54 4 2.11 Tooth 1 29 42.03 12 35.29 14 53.85 41 31.54 67 35.26 3 5 2 18 21 236 Vertebra 1 1.45 6 17.65 1 3.85 4 3.08 11 5.79 4 Rib 6 8.70 10 29.41 1 3.85 1 0.77 12 6.32 5 Scapula 0.00 0.00 0.00 2 1.54 2 1.05 Humerus 1 3 4.35 0.00 2 7.69 4 3.08 6 3.16 1 Radius 0 5 7.25 1 2.94 1 3.85 2 1.54 4 2.11 Ulna 2 2.90 0.00 0.00 2 1.54 2 1.05 2 1 Carpal 1 1.45 1 2.94 0.00 2 1.54 3 1.58 Metacarpal 2 2.90 0.00 2 7.69 15 11.54 17 8.95 1 1 Pelvis 1 1.45 2 5.88 0.00 5 3.85 7 3.68 Femur 1 1 1.45 0.00 1 3.85 5 3.85 6 3.16 1 Patella 0.00 0.00 0.00 0.00 0 0.00 Tibia 1 1 1.45 0.00 0.00 13 10.00 13 6.84 1 2 Tarsal 5 7.25 0.00 1 3.85 11 8.46 12 6.32 1 2 Metapodial 2 1 1.45 0.00 1 3.85 6 4.62 7 3.68 2 Metatarsal 1 1.45 0.00 0.00 13 10.00 13 6.84 1 1 1 Phalange 3 2 2.90 0.00 0.00 1 0.77 1 0.53 3 Sesamoid 2 2.90 0.00 0.00 1 0.77 1 0.53 Carapace plate 25 0.00 0.00 0.00 0.00 0 0.00 Total 33 1 69 100 34 100 26 100 130 100 190 100 3 11 6 28 21 253 Cranial 0 1 35 50.72 14 41.18 16 61.54 43 33.08 73 38.42 3 5 2 18 21 243 Cranial (excluding tooth) 0 0 6 8.70 2 5.88 2 7.69 2 1.54 6 3.16 0 0 0 0 0 7 Vert. þ rib 0 0 7 10.14 16 47.06 2 7.69 5 3.85 23 12.11 0 0 0 0 0 9 Girdle 0 0 1 1.45 2 5.88 0 0.00 7 5.38 9 4.74 0 0 0 0 0 0 Total Axial 0 0 8 11.59 18 52.94 2 7.69 12 9.23 32 16.84 0 0 0 0 0 9 Upper appendic. 2 0 4 5.80 0 0.00 3 11.54 9 6.92 12 6.32 0 1 1 0 0 0 Intermed. appendic. 1 0 8 11.59 1 2.94 1 3.85 17 13.08 19 10.00 0 3 0 3 0 0 Lower appendic. 2 0 4 5.80 0 0.00 3 11.54 34 26.15 37 19.47 0 1 1 4 0 1 Carpalþtarsalþphal. 3 0 10 14.49 1 2.94 1 3.85 15 11.54 17 8.95 0 1 2 3 0 0 Forelimb 1 0 13 59.09 2 50.00 5 71.43 27 36.49 34 40.00 0 2 2 2 0 0 Hindlimb 2 0 9 40.91 2 50.00 2 28.57 47 63.51 51 60.00 0 4 2 3 0 1 moderate degree of fragmentation can also be observed in relation Even though the number of bones with carnivore damage is to the degree of shaft preservation, since 54% and 66% of diaphyses limited, tooth mark are more abundant on high meat yield skeletal in VM4-I and VM4-II preserve over 50% of the shaft circumference. parts, such as upper limb bones. At VM4-II, all anatomical portions Limb bone length index also reflects a moderate degree of frag- exhibit tooth marks, though they are more frequent on upper limb mentation, since over 21% of bones from VM4-I and 36% from VM4- diaphyses. This pattern can be observed for Sizes 2 and 3 at VM4-I II still have over 50% of their expected length. Lastly, around 30% of and among Sizes 2, 3, and 5 at VM4-II (Table 10). shaft remains show green fractures, but dry fractures of diagenetic Furrowing consists in the deletion of bone portions due to origin are still more common in the VM4 assemblage (Table 6). carnivore ravaging. This alteration has been documented on 13 In relation to carnivore alterations, we have identified tooth bones from VM4-I and 27 bones from VM4-II, spanning a wide marks (Tables 7e9), furrowing (Table 11) green fractures associated range of anatomical parts, including compact bones, mandibles, with notches (Tables 12 and 13) and taphotypes (Table 14). and primarily on limb bones (Table 11; Figs. 2c, 3f and 3h). Despite Tooth marks have been observed on 4.4% of bones with well- the low numbers, it complements the tooth mark distribution data preserved cortical surfaces at VM4-I and 1.9% at VM4-II (Table 7). for a more accurate depiction of carnivore alteration of the faunal Only 5% of tooth-marked bones exhibit more than 5 tooth marks assemblage, since there are some bones that show furrowing even per bone, indicating a relatively low intensity of carnivore ravaging though no tooth marks are found on their cortical surfaces, such as (Table 8). Regarding the pit to score ratio, there is a clear pre- among some upper limb bones from Size 5 species (see Suppl. Data, dominance of pits over scores (Table 7). Other alterations, such as SF5 and SF7; and Fig. 2e). Furthermore, their distribution fre- digested bones (n ¼ 2) or bones polished due to saliva damage quencies can reach high levels for some size classes. For example, (n ¼ 6), are minimally present. the percentage of furrowing on Size 3b humeri (20%) or on Size 5 At VM4-1, medium-sized (equids) and small-sized species, such limb bones (100%) at VM4-II is notable (see Suppl. Data, SF7 and as Metacervoceros, present tooth marks. At VM4-II, in addition to SF8). these species, Soergelia and very large sized animals, such as ele- In relation to taphotypes, which characterise the nature and phants, rhinoceroses, and hippopotamuses, also show tooth marks distribution of furrowing, there is a predominance of Type 15 (both (Table 8). All the remains with tooth marks belong to adult in- epiphyses deleted), with 33% at VM4-I and 35% at VM4-II (Table 14; dividuals, except for an immature Metacervoceros and a juvenile Fig. 3h). Most bones presenting a Type 15 taphotype are upper Mammuthus, both from VM4-II. limbs, such as humeri and femora, which have a lower density than The anatomical distribution of tooth marks is outlined in Table 9. intermediate and lower limbs. This characteristic makes them more At VM4-I, tooth marks are found on limb bones, compact bones vulnerable to carnivore ravaging. In contrast, denser elements, such such as astragali or calcanei, as well as axial bones, such as pelves. as metapodials and tibiae, are more complete, even exhibiting type 8 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 5 Skeletal part profiles at VM4-II in relation to MNE. MNE VM4-II 0 1 2 % 3 % 3a % 3b % Total 3 % 4 5 Carniv 1 Carniv 2 Carniv 3 indet Horn-antler 4 2.30 1 4.76 0.00 1 0.42 2 0.65 Cranial 2 1.15 0.00 0.00 0.00 0 0.00 1 15 Maxillar 1 0.57 0.00 1 2.13 5 2.10 6 1.96 2 Mandible 2 12 6.90 0.00 4 8.51 3 1.26 7 2.29 6 Teeth 9 9 98 56.32 3 14.29 33 70.21 109 45.80 145 47.39 2 4 4 57 21 305 Vertebra 2 8 4.60 4 19.05 0.00 10 4.20 14 4.58 5 Rib 4 1 10 5.75 7 33.33 0.00 1 0.42 8 2.61 1 13 Scapula 2 1.15 1 4.76 0.00 2 0.84 3 0.98 1 Humerus 5 4 2.30 0.00 0.00 9 3.78 9 2.94 2 5 Radius 5 2.87 2 9.52 1 2.13 9 3.78 12 3.92 1 3 1 Ulna 3 1.72 0.00 0.00 3 1.26 3 0.98 1 4 Carpal 2 1.15 0.00 0.00 3 1.26 3 0.98 1 1 1 Metacarpal 2 1.15 0.00 0.00 29 12.18 29 9.48 5 6 Pelvis 1 1 0.57 1 4.76 0.00 5 2.10 6 1.96 Femur 1 0.00 0.00 2 4.26 4 1.68 6 1.96 1 Patella 0.00 0.00 0.00 0.00 0 0.00 1 Tibia 2 1 2 1.15 0.00 2 4.26 7 2.94 9 2.94 1 4 1 Tarsal 6 3.45 0.00 1 2.13 19 7.98 20 6.54 1 4 Metapodial 3 4 2.30 2 9.52 0.00 6 2.52 8 2.61 16 Metatarsal 5 2.87 0.00 1 2.13 8 3.36 9 2.94 1 Phalange 2 3 1.72 0.00 2 4.26 2 0.84 4 1.31 23 1 Sesamoid 0.00 0.00 0.00 3 1.26 3 0.98 10 2 Carapace plate 3 0.00 0.00 0.00 0.00 0 0.00 Total 32 13 174 100 21 100 47 100 238 100 306 100 2 20 4 140 26 342 Cranial 9 11 117 67.24 4 19.05 38 80.85 118 49.58 160 52.29 2 4 4 64 23 320 Cranial (excluding tooth) 0 2 19 10.92 1 4.76 5 10.64 9 3.78 15 4.90 0 0 0 7 2 15 Vert. þ Rib. 6 1 18 10.34 11 52.38 0 0.00 11 4.62 22 7.19 0 1 0 0 0 18 Girdle 1 0 3 1.72 2 9.52 0 0.00 7 2.94 9 2.94 0 0 0 0 1 0 Total Axial 7 1 21 12.07 13 61.90 0 0.00 18 7.56 31 10.13 0 1 0 0 1 18 Upper appendic. 6 0 4 2.30 0 0.00 2 4.26 13 5.46 15 4.90 0 3 0 5 0 0 Intermed. appendic. 2 1 10 5.75 2 9.52 3 6.38 19 7.98 24 7.84 0 4 0 11 1 1 Lower appendic. 3 0 11 6.32 2 9.52 1 2.13 43 18.07 46 15.03 0 6 0 22 0 0 Carpalþtarsalþphal. 2 0 11 6.32 0 0.00 3 6.38 27 11.34 30 9.80 0 2 0 38 1 3 Forelimb 5 0 18 56.25 3 75.00 1 25.00 55 56.12 59 55.66 0 10 0 19 1 2 Hindlimb 4 1 14 43.75 1 25.00 6 150.00 43 43.88 50 47.17 0 5 0 8 1 0 Table 6 General taphonomic alterations in VM4 site. *% respect to total amount of specimens. **% with respect to long bones. ***% respect to total amount of specimens excluding teeth. VM4-I % VM4-II % All sample 1732 3166 NISP excluding teeth 1272 2459 Fragmentation Bone length <3 cm* 742 42.8 1142 36.1 Bone length 3.1 cme5cm* 436 25.2 930 29.4 Bone length 5.1e9.9 cm* 359 20.7 647 20.4 Bone length >10 cm* 195 11.3 447 14.1 Long bone fracture NISP of long bone with green fracture** 46 23 30 21 NISP of long bone with dry fracture** 118 59 31 21.7 NISP of long bone with green & dry fracture** 18 9 12 8.3 NISP of long bone with indet fracture** 19 9 70 49 Circumference NISP shaft circumference <50%** 89 46.4 40 34.2 NISP shaft circumference >50%** 32 16.7 24 20.5 NISP shaft circumference >75e100%** 71 37.0 53 45.3 Length respect to total long bone NISP % length respect the total of long bone >50%** 157 78.5 84 63.6 NISP % length respect the total of long bone <50%** 13 6.5 25 18.9 NISP % length respect the total of long bone 75e100%** 30 15.0 23 17.4 Alterations NISP good surface preservation, excluding tooth*** 965 75.9 2073 84.3 NISP weathering 1e2 stage* 137 7.9 224 7.1 NISP weathering 3e4 stage* 0 0.0 0 0.0 NISP with concretion* 151 8.7 188 5.9 NISP manganese or oxide staining* 287 16.6 319 10.1 Abrasion NISP with abrasion* 516 29.8 706 22.3 NISP light abrasion* 157 9.1 212 6.7 NISP moderate abrasion* 210 12.1 302 9.5 NISP intense abrasion* 149 8.6 188 5.9 NISP with biochemical* 304 17.6 354 11.2 9 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 7 Tooth mark frequencies. Tooth mark type and number of marks per bone. *: NISP frequencies excluding teeth and bones with poor surface preservation; **: Frequencies in relation to bones with tooth marks. NISP excluding tooth and bones with bad surface preservation VM4-I % VM4-II % 965 2078 Bones with tooth marks 42 4.4* 40 1.9* Bones with pits 30 71.4** 26 65.0** Bones with scores 8 19.0** 5 12.5** Bones with pit and scores 4 9.5** 8 20.0** Bones with punctures 0 0.0** 1 2.5** Nº tooth mark for specimen (5-10) 2 4.8** 2 5.0** Nº tooth mark for specimen (2-4) 20 47.6** 26 65.0** Nº tooth mark for specimen (1) 20 47.6** 12 30.0** Table 8 Distribution of bones with tooth marks (TM) per species. Level VM4-I VM4-II NISP excluding teeth and bones with NISP with TM %TM NISP excluding teeth and bones with NISP with TM %TM bad surface preservation bad surface preservation M.meridionalis 3 1 33.3 S. etruscus 3 0.0 10 2 20.0 H. antiquus 16 0.0 1 1 100.0 Bovinae 6 1 16.7 10 0.0 E. altidens 48 8 16.7 104 4 3.8 M. rhenanus 14 2 14.3 22 2 9.1 P. verticornis 11 1 9.1 25 4 16.0 S. minor 5 0.0 6 2 33.3 Herbivore indet. Size 2 47 1 2.1 60 3 5.0 Herbivore indet. Size 3 40 5 12.5 69 4 5.8 Herbivore indet. Size 3a 22 3 13.6 6 2 33.3 Herbivore indet. Size 3b 109 18 16.5 71 11 15.5 Herbivore indet. 544 3 0.6 1498 4 0.3 Table 9 Table 10 Anatomical distribution of bones with tooth marks in VM4-I and VM4-II. Frequencies of tooth marks respect bones with good cortical surface according to appendicular limbs where upper: Humerus þ Femur; intermediate: Anatomical distribution of tooth marks Radius þ Tibia þ Ulna); and lower (metapodials). Size 3 included (3a, 3b and 3). VM4-I VM4-II Animal Size VM4-I VM4-II Animal Size 2 3 3a 3b Indet. 1 2 3 3a 3b 5 Indet. 2 3 2 3 2 3 5 2 3 5 Mandibles 1 1 Ribs 1 NISP % tooth NISP with % Tooth marks Vertebrae 1 with marks Tooth Scapulae 1 Tooth marks Humeri 1 1 3 1 5 2 marks Radii 1 2 Upper limbs 1 6 25 46.2 1 6 2 16.7 31.6 66.7 Pelvises 1 2 Intermediate limbs 10 0 40 2 0 8.6 0 Femurs 1 1 1 Lower limbs 2 0 7.7 1 2 1 20 3.7 20 Tibiae 10 Metacarpals 1 2 1 Metatarsals 1 2 Tali 1 1 Calcanei 1 1 1 (Types 2, 3b, 4 and 5; Fig. 4b and c); and then by micro-notches and Shafts 1 1 7 3 1 3 Indet. 3 1 4 1 3 1 3 1 pseudo-notches. At VM4-II, notches are also mostly found among Total 3 5 3 28 3 1 6 4 2 19 4 4 Size 2e3 animals (Table 12), although micro-notches and pseudo- notches predominate in this level, followed by single notches, with double and opposing notches being less well represented. 0 taphotypes, which means that they are complete (Table 13; Fig. 2a) Proximal epiphyseal deletion has also been observed in over 4. Discussion 36% of limb bones at VM4-II and 48% at VM4-I (taphotypes 2e7), while several examples of distal epiphysis deletion corresponding 4.1. Comparing VM4-I and VM4-II to taphotypes 9e12 have also been documented (Table 14; Fig. 2a). Regarding bone fracture angles, there is a predominance of In this paper we have presented the first comprehensive anal- angles with values between 75 and 110 among bones with green ysis of the faunal assemblage retrieved from VM4 during the 2005, fractures (Table 13). 2013e2015 and 2017e2019 field seasons. Contrary to Palmqvist At VM4-I, notches have only been documented among Size 2 et al. (2022), there are no data from the 2020 field season as no and 3 animals (Table 12). Single notches comprise 47% of the excavations were carried out at VM4 that year. sample (Figs. 3 and 4d), followed by double or opposing notches As outlined in previous works (Granados et al., 2021; Luzo n 10 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 11 Bones exhibiting furrowing from VM4. VM4-I Bovinae Equus Praemegaceros Indet. Talus 3 Femur 1 1 Humerus 1 1 2 Tibia 3 Pelvis 1 1 1 Total 1 6 2 4 VM4-II Stephanorhinus Hippopotamus Mammuthus Bison Equus Praemegaceros Sorgelia Indet Mandible 2 1 1 Humerus 2 1 3 Femur 1 3 1 Radio 1 2 Tibia 2 1 Pelvis 2 Calcaneous 1 Carpal-tarsal 1 Indet. 2 Total 3 1 1 2 8 1 2 9 Table 12 the formation of VM4-II, an interpretation that would be supported Notches identified at VM4. Type 1: complete single notches; Type 2: double by the geochemical data discussed in Granados et al. (2021). opposing complete notches; Type 3a: incomplete notches where inflection points However, more data from tooth wear, archaeobotanical remains are missing; Type 3b: complete and other incomplete notches where one inflection points is missing; Type 4: double overlapping notches; Type 5: a notch and other and stratigraphic micromorphology are needed for a more accurate double opposing notches; Type 6: pseudo notch; Type 7: micronotch (< 1 cm) (See depiction of the palaeoenvironmental changes that seem to have Suppl. Data, Fig. 1). occurred between the formation of VM4-I and VM4-II. VM4-I % VM4-II % Total % From a taphonomic perspective, VM4-I and VM4-II faunal as- semblages share several similarities, since both are characterised Type 1 8 20.0 13 16.7 21 17.8 Type 2 3 7.5 4 5.1 7 5.9 by good preservation conditions, reflected in a limited incidence of Type 3a 11 27.5 16 20.5 27 22.9 weathering. While carnivore action is present in both levels, there Type 3b 5 12.5 4 5.1 9 7.6 is a greater incidence at VM4-I than at VM4-II. At VM4-I there is a Type 4 2 5.0 4 5.1 6 5.1 greater degree of fragmentation, as shown by the proportion of Type 5 2 5.0 3 3.8 5 4.2 bones smaller than 3 cm (43% NISP < 3 cm in VM4-I vs. 36% NISP < Type 6 1 2.5 10 12.8 11 9.3 Type 7 8 20.0 24 30.8 32 27.1 3 cm in VM4-II) and the greater proportion of bones with shaft Total 40 78 118 circumference preservation under 50% (46% in VM4-I vs. 34% in VM4-II) (Table 6). There is also a greater degree of skeletal bias at VM4-I, with 8 elements per individual among Size 2 animals, 17 Table 13 elements among Size 3 animals, and 1.5 elements among Size 5 Descriptive statistics for green fracture angles on bones from size 3 animals, based animals; by contrast, at VM4-II the values are higher, with 17 ele- on the nature of the fracture plane: oblique and longitudinal. ments per individual among Size 2 animals, 16 elements per indi- VM4-I VM4-II vidual among Size 3 animals and 3.3 elements per individual Oblique Longitudinal Oblique Longitudinal among Size 5 animals. This skeletal bias may be related to the greater proportion of carnivore damage observed within the VM4-I Average < 90 76.6 74.7 77.3 76 faunal assemblage, as noted in Table 7. 4.4% of bones from VM4-I Average > 90 99.7 102.6 104.5 102 show tooth marks versus 1.9% at VM4-II (Table 7). In relation to bone fracture angles, no appreciable differences are observed be- tween the two levels (Table 13), although they are consistent with et al., 2021), geoarchaeological and taphonomic evidence proves ntara García et al., 2006; carnivore bone breakage patterns (Alca the existence of at least two palaeosurfaces at VM4, resulting from Suppl. Data, SF7), particularly in relation to canid and hyaenid at least two depositional events. Accordingly, it is relevant to start n et al., 2019). Both levels also exhibit feeding behaviour (cf. Mocla by outlining the similarities and differences between both sub- similar proportion of Type 15 taphotypes (shaft cylinders; Table 14). levels. The species represented at VM4-I and VM4-II are essentially Expanding the sample size will provide additional data to assess identical, except for Hemibos and Megantereon, which are only the extent to which these differences between sublevels are present in VM4-II. The proportion of herbivores and carnivores on replicated or ought to be discarded. both levels is remarkably similar (70/30 in VM4-I vs. 69/31 in VM4- II). Among the herbivore taxa, Cervidae (Metacervoceros and Prae- megaceros) is the best represented family in both sublevels, 4.2. Relationship and comparative assessment between VM4 and although the most abundant species is E. altidens. VM3 Despite their similarities, there are some slight differences in the assemblage composition that might indicate some sort of palae- Comparing our results with those provided by Arribas and oecological change in the landscape of Venta Micena. In VM4-II, Palmqvist (1998, 1999), Palmqvist and Arribas (2001), Espigares species associated with open habitats, such as horses and rhinos, (2010) and Palmqvist et al. (2022), it is true that VM3 and VM4 are less frequent than in VM4-I, with a corresponding increase in exhibit some similarities (Table 16). However, there are also clear the proportion of cervids from 28% of MNI in VM4-I to 37% in VM4- differences (Table 16 and 17). II. These variations may indicate a more humid environment during The first difference is in the origin of the materials. We know the 11 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 14 Taphotypes identified at VM4, following the criteria outlined by Domínguez Rodrigo et al. (2015). VM4-I Taphotype 0 1 3 4 5 6 9 10 11 12 15 Femur 2 Humerus 1 2 3 Metatarsal 1 Radio-ulna 1 Tibia 1 5 1 1 Total 2 5 1 3 1 6 % 11.1 27.8 5.55 16.7 5.55 33.3 VM4-II Taphotype 0 1 3 4 5 6 9 10 11 12 15 Femur 1 1 4 Humerus 1 1 2 3 Metacarpal 1 1 Metatarsal 1 Radio 1 1 1 1 2 Tibia 1 2 1 1 1 Total 3 1 3 2 3 2 1 1 2 10 % 10.71 3.57 10.71 7.14 10.71 7.14 3.57 3.57 7.14 35.71 Fig. 3. Bones from VM4 compared with bones retrieved from actualistic open-air sites. 4a: Horse mandible with furrowing on the mandibular ramus. 4b-c: neotaphonomic ex- amples of mandibles ravaged by wolves at a kill site in El Campelo (Galicia). 4d: distal humerus epiphysis from VM4. 4e: distal humerus epiphysis from a horse hunted by wolves at El Campelo (Galicia). 4f: proximal ulna epiphysis with furrowing. 4g: similar feeding sequence documented by wolves at a kill site in El Campelo (Galicia). 4h: two equid shaft cylinders on a tibia and a metatarsal from VM4. 4i-k: immature equid femoral shafts from El Campelo (Galicia). 12 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Fig. 4. Tooth-marked bones from VM. 4a: Scapula with a puncture; 4 b-d: long bones with different notch types. 4e: pelvis with biochemical alterations. stratigraphic and spatial provenance for the VM4 sample: 39 m2 treated with caution. Despite the above, we will compare our re- located at 552462 E, 4176355 N. On the contrary, the provenance of sults for VM4 with those previously published for VM3 because up the VM3 sample is heterogeneous, with almost half of the materials to this point it is the major reference when it refers to VM. Any lacking a clear origin. Espigares ph.D. dissertation (2010) (super- future taphonomic study involving VM assemblages must be car- vised by B. Martínez-Navarro and P. Palmqvist) states that 3603 of ried out on remains with known provenance and stratigraphic 7171 elements come from known areas (Table 15). The remaining position, because the provenance of several samples published by 3568 elements come from indeterminate localities (Espigares, Espigares (2010) and Palmqvist et al. (2022) remains unknown. 2010: 114, Table 7.2.). Moreover, for the remains coming from Taxonomic representation also differs between VM3 and VM4, known emplacement, 723 elements come from fossil outcrops even if their taxonomic composition is similar (Luzo  n et al., 2021; other than VM3. These fossils account for 20% of the known prov- Arribas and Palmqvist, 1998, 1999; Palmqvist and Arribas, 2001). enance material and 10% of the total. The significance of this lies in According to MNI values, equids are the most represented herbi- the fact that Palmqvist et al. (2022) have based their study on the vores at VM3 (35%), a higher frequency than in VM4-I (25%) and materials studied by Espigares (2010), mistakenly attributing them VM4-II (22.5%). Megafaunal remains (elephants, rhinos and hippos) all to VM3 (Palmqvist et al., 2022: p. 4, Table 2). are more abundant at both VM4-I (25%) and VM4-II (15%) than at To sum up, around 50% of the remains cannot be associated with VM3 (8%) (Tables 2 and 3). However, other taxa show similar values any specific VM sector, and 20% of the subsample with known in VM3 and VM4. Cervids represent 28.6% of the MNI at VM4-I and origin does not come from VM3. The VM3 fossil assemblage sensu 28.3% at VM3, increasing to 37% at VM4-II. Caprids also have similar stricto is thus reduced to 2880 specimens (Espigares, 2010; frequencies in the VM3 assemblage (13.6%) to those found at VM4-I Table 15; Fig. 5). For those reasons, until this smaller sample is (14.3%) and VM4-II (10%). In summary, very large mammals are restudied separately, all previously reported results should be more frequent at VM4, while equids are better represented at VM3. 13 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Table 15 Palmqvist and Arribas (2001), Palmqvist et al. (2011) and Palmqvist Number of studied remains by Espigares, (2010: Table 7.2), with a breakdown of et al. (2022); meanwhile, only adult individuals are represented at their source provenance. “Corte”: large-scale excavation; “Sondeo”: test-pit. VM4-I and VM4-II (Tables 2 and 3). Location Number of remains % A direct correlation between the size of the species and the Corte I 403 5.62 proportion of non-adult specimens at VM3 has been claimed Corte II 51 0.71 repeatedly (Palmqvist et al. 1996; Arribas and Palmqvist, 1998; Corte III 2.880 40.16 Palmqvist et al., 2022). Applying the same linear regression method Corte IV 16 0.22 employed by Palmqvist et al. (2022), we note that no such corre- Corte V 1 0.01 Sondeo I 199 2.78 lation between size class and proportion of non-adult individuals is Sondeo II 33 0.46 observed for the faunal assemblages of VM4-I, VM4-II, VM4 (this Sondeo III 12 0.17 study) or VM4 [in Palmqvist et al. (2022)]. This lack of correspon- Sondeo IV 2 0.03 dence further reinforces the notion that VM3 and VM4 constitute Sondeo V 6 0.08 Indeterminate 3.568 49.76 different locations in a complex landscape, without evidence at VM4 of the same prey selection strategies proposed for VM3 (Table 17). Moreover, in terms of skeletal part profiles, Arribas and Palmqvist (1998), Palmqvist and Arribas (2001), Espigares (2010) There are also some differences in the carnivore palaeoguild and Palmqvist et al. (2011) propose that smaller animals (Size 2), represented at both localities. The giant hyaena P. brevirostris con- such as S. minor, C. alba or M. rhenanus, were transported as com- stitutes 34% of carnivore MNI at VM3 (Arribas and Palmqvist, 1998, plete carcasses to VM3, while medium-sized animals (Size 3), such 1999; Palmqvist and Arribas, 2001), a higher value than those re- as equids, would have been the subject of a more specialised ported for VM4-I (22.2%) and VM4-II (16.7%). This trend become strategy, characterised by the preferential transport of appendic- even more striking when we assess their representation using NISP ular elements. In contrast, we note that both Size 2 and Size 3 an- values (26.9% at VM3 vs. 3.2% at VM4-II). P. brevirostris is decidedly imals show a similar pattern of skeletal part representation at VM4- more abundant at VM3. I and VM4-II (Tables 4 and 5). All skeletal portions are represented Mortality profiles constitute a third area in which VM3 and VM4 for both size classes and they show similar biases in skeletal part differ, particularly regarding non-adult age groups. At VM4-I and proportions, which suggests a lower incidence of preferential prey VM4-II, only P. verticornis and the very large taxa exhibit non-adult selection than at VM3. values over 45%, while all medium, large, and very large-sized an- The notion of a less selective depositional strategy is supported imals at VM3 exhibit non-adult values over 45%. A borderline by the existence of several joints, ribs, and vertebrae in natural example are equids, of which there are 46% of non-adult individuals anatomical connection for Size 2, 3, and 5 animals, as discussed at VM3 according to Arribas and Palmqvist (1998, 1999), Palmqvist here and previously described by Luzo  n et al. (2021) and Palmqvist and Arribas (2001) and Palmqvist et al. (2011), or 55% according to et al. (2022, Fig. 6), while only cranial and lower appendicular el- Palmqvist et al. (2022). In contrast, their percentage decreases to ements are found in anatomical connection at VM3 (Palmqvist 28.7% at VM4-I and 33% at VM4-II. These divergences in terms of et al., 2022). The pattern observed at VM4 suggests that some mortality profiles increase when we assess the data reported for carcasses accumulated on the site itself, without experiencing any Pachycrocuta brevirostris, since 40e50% of MNI corresponds to non- transport by large carnivores or other biological agents as inferred adults at VM3, according to Arribas and Palmqvist (1998, 1999), for VM3. Table 16 Summary of differences reported for the two main Venta Micena sites (VM3 and VM4). Features Similarities Differences Sample provenance The published VM3 samples are heterogeneous in origin; VM4 sample is well recorded and spatially located. Sedimentological Similar sedimentological processes evidence Stratigraphy Similar strata present at VM3 and VM4 Two sublevels have been identified at VM4 Biostratigraphy Similar species Water transport Absent in both Weathering More prolonged subaerial exposure at VM3 Very fast burial rate at VM4 Herbivore frequencies Size 5 animals comprise 25e17.5% of MNI at VM4 vs. 6% at VM3 Carnivore frequencies Pachycrocuta comprises 34.6% of carnivore MNI at VM3 vs. 22.2% at VM4-I and 16.7% at VM4-II Mortality Profiles VM3 >45% non-adults in Size 3 and 5 animals Herbivores VM4 >50% non-adults in Size 5 animals VM4 <33% non-adults in Size 2 and 3 animals Pachycrocuta mortalitiy VM3 50% non-adults profile 0% of non-adults at both VM4-I and VM4-II Skeletal part profiles Anatomical bias VM3: Size 3 animals are preferentially transported selective, size 2 animals are complete VM4: Both Size 2 and 3 are similarly represented. Anatomical connections VM3: Only cranial bones and basipodial elements VM4: Abundant (including axial and appendicular elements) Tooth marks VM3: High frequencies (29.4%; Arribas, 1999) VM4: Low frequencies (4.4% at VM4-I, 1.9% at VM4-II) Bone fractures VM3: Predominance of green fractures over dry fractures. VM4: Predominance of dry fractures over green fractures. Taphotypes Presence of bone cylinders (Type 15) at both sites. 14 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031  n et al., somewhat different assemblage formation processes (Luzo 2021; Granados et al., 2021). 4.3. Site function at VM4 After establishing that VM3 and VM4 are the result of different site formation processes, we now need to discern the functionality of VM4, which we will interpret through the data presented in the Results section. As a starting point, we can confirm that VM4 is not an accu- mulation generated by felids, canids, or ursids, since, apart from foxes, canids are generally agents of dispersal rather than accu- mulation (see Yravedra et al., 2012). A similar pattern towards favouring dispersion is observed among ursids (Arilla et al., 2014). Felids, such as leopards, can generate bone accumulations (Brain, 1981; Ruiter and Berger, 2000), although their prey tends to be considerably smaller than most of the species represented at VM4; additional differences are observed in terms of the expected mor- tality and skeletal part profiles of felid-generated accumulation and those described for VM4. Other large felids, such as lions, only rarely produced bone accumulations, often due to exceptional cir- cumstances (Arriaza et al., 2019). Hyaenas are one of the main biological agents of bone accu- mulation at a wide range of Pleistocene sites. One of the core fea- tures of hyaenid den sites is a high frequency of tooth marks, often Fig. 5. Comparison among reported NISP values for VM3: grey circle, NISP (total) provided by Palmqvist et al. (2022); blue circle, NISP (total) provided by Espigares above 50%, a pattern observed at both modern assemblages (2010); green circle, NISP from known quarries/outcrops provided by Espigares (Bearder, 1977; Skinner et al., 1986; Blumenschine, 1988; Hill, 1989; (2010); orange circle, NISP from indeterminable location materials provided by Cruz-Uribe, 1991; Lam, 1992; Capaldo, 1995; Marean and Kim, 1998; Espigares (2010); yellow circle, NISP from VM3 provided by Espigares (2010); red Pickering, 2002; Lacruz and Maude, 2005; Faith, 2007; Egeland circle, NISP of known quarries/outcrops other than VM3 provided by Espigares (2010). et al., 2008; Prendergast and Rodrigo, 2008; Kuhn et al., 2009), as All circles are to scale. well as in Pleistocene contexts (Brugal et al., 1997; Marra et al., 2004; Michel, 2004; Andrews, 2008; Villa et al., 2010). While Carnivore alterations are also far more frequent at VM3 than at there are several examples of hyaenid accumulations with tooth VM4 (Arribas and Palmqvist, 1998; Espigares, 2010). Over 1555 mark frequencies falling below 50% (likely due to external factors remains (>30%) are described to exhibit Pachycrocuta tooth marks that can play a role in reducing the intensity of bone ravaging, such (Arribas, 1999; Espigares, 2010), whereas neither of the sublevel as reduced ecosystem competition) tooth marks are generally still assemblages at VM4 has more than 5% of remains with tooth marks. present in over 20% of bones (Faith, 2007; Pokines and Kerbis The last factor to consider in a comparative assessment between Peterhans, 2007; Kuhn et al., 2008; Fourvel et al., 2015; Mwebi VM3 and VM4 is the time span in the biostratinomic phase, the and Brugal, 2018; Linares-Mata s et al., 2021). Per contrary, the taphonomic stage between death and definitive burial. Weathering VM4 assemblage is far from those values, with barely 4.4% of bones rates, which are useful for understanding the degree of subaerial exhibiting tooth marks in VM4-I, and only 1.9% for VM4-II. Fur- exposure, are low for both levels of VM4 (Table 6), which would rowing also has a relatively low incidence (<6%) throughout the suggest a quick burial rate, favouring the preservation of the assemblage. anatomical connections documented at the site (Palmqvist et al., Tooth mark frequencies in Pachycrocuta-generated assemblages 2022). Palmqvist et al. (2022) argue that the proportion of are high (Saunders and Dawson, 1998; Espigares, 2010; Garrido abraded remains at VM4 reported by Luzo  n et al. (2021) (29.8% at et al., 2010; Madurell-Malapeira et al., 2017; Gibert and VM4-I and 22.3% at VM4-II, respectively) must be in error since no Ferrandez-Can ~ adell, 2015) to moderate (Mazza et al., 2004; abrading substances are present in the VM4 sediment. However, we Tappen et al., 2007; Bourguignon et al., 2016; Cheheb et al., 2019; highlight that the abrasion is predominantly light in nature and Coil et al., 2020). In either case, these frequencies are higher than could have been generated by the contact of suspended silt and clay the values reported by Palmqvist et al. (2022) (5.5%) and in this particles with bone surfaces in the context of low energy water study. The lower % of tooth marks in our study compared to flows during deposition (cf. Thompson et al., 2011). Palmqvist et al. (2022) may be due to the fact that no biochemical On the basis of all the differences observed between VM3 and marks were observed in the latter while 59 have been reported in VM4 (summarised in Tables 16 and 17), we argue that these two this study (e.g., Fig. 4e). A mistaken characterisation of biochemical neighbouring and penecontemporaneous sites experienced alterations as representing tooth marks has several precedents in Early Pleistocene taphonomy, including at Olduvai Bed I sites Table 17 Regression analysis using the log10 of estimated adult body mass (BM) of herbivore taxa at VM3 and VM4 as independent variable and log10 of non-adult individuals at each locality as dependent variable. The only slope with a p-value <0.05 for which the null hypothesis (H0: slope ¼ 0) is the linear regression of log immature on logBM at VM3. VM3 VM4 VM4-I (This study) VM4-II (This study) VM4 (This study) Palmqvist et al. (2022) Palmqvist et al. (2022) log%inmature R2 slope p-value R2 Slope p-value R2 slope p-value R2 slope p-value R2 slope p-value on logBM 0.694 0.340 <0.01 0.005 0.076 0.860 (ns) 0.060 -0.318 0.525 (ns) 0.329 0.779 0.106 (ns) 0.151 0.338 0.301 (ns) 15 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Fig. 6. Neotaphonomic examples of carcasses found at open-air sites. a-c: Wildebeest carcass hunted by lions and scavenged by hyaenas, with some bones in anatomical connection. d: Complete limb bones in anatomical connection from a wildebeest carcass. e-f: Exposed wildebeest carcass at a lion kill site. g: Wildebeest carcass at a kill site with basipodium in anatomical connection. h-o: Young horse carcasses at wolf kill sites; note the presence of articulated remains and complete limb bones. h, i: Pictures taken 1h after a wolf feeding bout, j: Picture taken 4 days after the kill, k-o: Pictures taken 1e2 months after the kill (see Yravedra et al., 2011, 2012 for more details pertaining the carcasses shown here). (Domínguez-Rodrigo and Barba, 2006). Even if we consider the bone specimen is consistently low (100% of bones with carnivore VM3 data reported by Espigares (2010) and Palmqvist et al. (2022) alterations exhibit <5 tooth marks: Table 7). This pattern is simply to be accurate (29% of bones with tooth marks), the values for this inconsistent with the values reported for hyaenid accumulations locality are considerably higher than those present at both VM4 (Brugal and Fosse, 2004; Domínguez-Rodrigo and Pickering, 2010). sublevels. Furthermore, we can discard the notion that VM4 tooth Several other lines of evidence further indicate that it is very mark values are underrepresented due to the incidence of other unlikely that any of the VM4 sublevels could be a Pachycrocuta den taphonomic processes causing high levels of bone fragmentation or site. There is an absence of immature Pachycrocuta remains etheir affecting the preservation of bone surfaces (Cruz-Uribe, 1991; presence is considered a strong indicator of hyaena denning ac- Brugal et al., 1997; Pickering, 2002; Kuhn et al., 2008), since cortical tivities, as reported for VM3 (Kruuk, 1972; Mills, 1989; Arribas and bone surfaces are very well preserved at both VM4 sublevels. Palmqvist, 1998; Brugal and Fosse, 2004); besides, there is a much Several additional circumstances that may influence the vari- lower proportion of digested bone remains at VM4 than at other ability observed in tooth mark frequencies within hyaenid accu- den sites (Kruuk, 1972; D'Errico and Villa, 1997; Marra et al., 2004; mulations include the presence of often underreported Michel, 2004; Brugal and Fosse, 2004; Villa et al., 2010; Gibert and “inconspicuous” tooth marks caused by young individuals Ferrandez-Can ~ adell, 2015). The pit/score ratio, with the clear pre- (Domínguez-Rodrigo and Pickering, 2010; Linares-Mata s et al., dominance of the former over the latter at VM4 (Table 7), also 2021), the number of feeders, (Fourvel et al., 2015), bone density, diverge from the usual expectations and values reported for or prey size, whereby the bones of larger animals tend to exhibit hyaenid-accumulated assemblages, where either scores predomi- more tooth marks (Faith, 2007; Kuhn et al., 2009). This prey size- nate or scores exhibit similar values to pits (Egeland et al., 2008; dependent pattern is indeed replicated at VM4-I and VM4-II Domínguez-Rodrigo et al., 2012; Arriaza et al., 2019). The low (Tables 8 and 9), although the overall number of tooth marks per incidence of furrowing at VM4-I and VM4-II is reflected in the large 16 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 amount of epiphyses found at the site. The presence of bones in for accumulations generated by felids such as leopards (Ruiter and anatomical connection and the green bone fragmentation patterns Berger, 2000) or lions (Arriaza et al., 2016) but not those for hyaenid in limb bones (23%; Table 6) reported for VM4 do not match the den assemblages. patterns observed for hyaenid-accumulated assemblages (Kruuk, Based on the contextualisation of the results discussed above, it 1972; Cruz-Uribe, 1991; Brugal and Fosse, 2004; Domínguez- is very plausible to argue that VM4 constitutes an open-air locality Rodrigo and Pickering, 2010; Mwebi and Brugal, 2018). Lastly, the where multiple agents and depositional processes were at play, VM4 mortality profiles for Size 3 animals, with a predominance of leading to a complex set of accumulation episodes, which may also adult individuals, also contrast to hyaenid-accumulated assem- include some instances of natural deaths. The hunting and scav- blages, which show a higher number of non-adult specimens enging of herbivore carcasses by carnivores are reflected in the (Kruuk, 1972; Cruz-Uribe, 1991; Brugal and Fosse, 2004; Arribas, presence of tooth marks on bone surfaces and the biases of the 1999; Domínguez-Rodrigo and Pickering, 2010; Linares-Mata s skeletal part profiles in the absence of any substantial water et al., 2021). transport (Tables 2e5). It is not implausible that some of the Size 5 Other accumulation dynamics at open-air sites resemble more animals (c.20% of MNI at VM4) present at the site, namely ele- closely than a den site the patterns observed at VM4. Skeletal parts phants, rhinoceroses, and hippopotamuses, may correspond to in- in anatomical connection, complete bones, and low tooth mark dividuals killed by large sabre-toothed felids, who are considered to frequencies are commonly found at these sites (Sept, 1994; have targeted megafaunal species quite consistently (e.g. Palmqvist Coumont, 2009; Hutson and Cain, 2008; Yravedra et al., 2013). et al., 2003, 2008a, b). Alongside the predominance of non-adult Lion-hunted carcasses exposed to the feeding behaviour of other individuals among Size 5 species, there are other lines of evi- scavengers, such as hyaenas and jackals, tend to exhibit complete dence that support the notion that felids played a relevant role in limb bones and axial elements in anatomical connection (Fig. 6). the accumulation and alteration of the faunal assemblage. The pit/ Domínguez-Rodrigo et al., 2007 have observed that hyaenid score ratio more closely resembles the taphonomic patterns of fe- ravaging is more intense at den sites than at open air sites, which lids than those of hyaenids (Pickering et al., 2011; Arriaza et al., correspondingly have lower tooth mark frequencies. Neo- 2019) and the number of tooth marks per bone is lower than 5 taphonomic research on wolf kill sites show that bones present for VM4-II and lower than 7 tooth marks for VM4-I (see Suppl. Data, different degrees of anatomical connectivity depending on the SF8), which fits more closely the patterns described for felids than nature and intensity of consumption (Fig. 6h-o). We have observed those described for hyaenids or canids, which can leave more than that axial remains and limb bones tend to preserve a considerable 10 tooth marks per bone (Yravedra et al., 2011; Domínguez-Rodrigo degree of anatomical connection when wolves engage with the et al., 2012) with the exception of lycaons (Yravedra et al., 2014). hunted carcasses of young horses for just one feeding bout (Fig. 6d In addition to felids, hyaenids also interacted with carcasses at and h), retaining those connections for several months after such VM4, likely acting as scavengers; their activity is illustrated by the consumption episodes (Fig. 6k-o). Foxes also tend to leave behind presence of diaphysiary cylinders on limb bones from Size 3 adult articulated remains, with lower tooth mark frequencies and bone individuals (Fig. 2; see also Sutcliffe, 1970; Cruz-Uribe, 1991; Fosse, fragmentation at open air sites than at den sites (Krajcarz and 1994; Pokines and Kerbis Peterhans, 2007; Fosse et al., 2011; Krajcarz, 2014; Yravedra et al., 2014; Young et al., 2015; Arilla Domínguez-Rodrigo et al., 2012, 2015). Other carnivores, such as et al., 2019). jaguars, can crush and collapse some bone epiphyses but do not At some open-air waterholes, Hutson (2012) noted very high produce such diaphysiary cylinders (Domínguez-Rodrigo et al., tooth mark frequencies (c.50%) during the dry season, although 2012; Rodríguez-Alba et al., 2019). Lions are also not known to most tooth marks are found on axial remains, particularly vertebrae produce these alterations on the limb bones of medium- or large- and ribs. Arilla et al. (2019) have also shown that girdle and axial sized species (Gidna et al., 2013; Pobiner and Blumenschine, bones tend to present more tooth marks than limb bone diaphyses 2003). Wolves have been observed to generate shaft cylinders at open air sites. However, the lower density of these remains in when consuming deer bones or immature horse individuals (Fig. 4) comparison to bone shafts and other anatomical parts (Lam et al., or collapse some epiphyses of adult horses (Fig. 2; Yravedra et al., 1998, 1999) often leads to their disappearance from the fossil re- 2011) or even bison (Fosse et al., 2012), but they generally do not cord due to a wide range of potential density-mediated attritional generate diaphysiary cylinders on large adult prey. The presence of processes, which may explain why fossil assemblages from open- these taphonomic alterations at both VM4-I and VM4-II (Table 14) air sites tend to have low tooth mark frequencies (Hutson and can more likely be attributed to the feeding behaviour of P. brevir- Cain, 2008). ostris, which may have even hunted some of these carcasses, a While carnivore predation often occurs at open-air sites, natural proposition which cannot be ruled out either (Turner and Anto  n, mortality can also lead to the accumulation of faunal remains in 1996; Dennell et al., 2008; Vinuesa et al., 2016; Iannucci et al., these localities (Haynes, 1981, 1987, 1988, 1991). Ecological and 2021). neotaphonomic studies have noted how carnivore feeding behav- Alongside felids and hyaenids, it is possible that other carnivores iour, such as the intensity of carcass consumption and bone also intervened in the modification of the VM4 faunal assemblage. ravaging, as well as scavenging opportunities, are influenced by After comparing the fracture angles at the site with the patterns seasonality and associated factors, such as the levels of carnivore generated by hyaenids and canids (cf. Mocl an et al., 2019), it seems competition or resource limitation (Blumenschine, 1986; Tappen, that fracture angles at VM4 are closer to those left by canids, an 1992; Sinclair et al., 2003; Pereira et al., 2014; Arilla et al., 2019; observation that would also be consistent with the taxonomic Linares-Mata s and Yravedra, 2021; Linares-Mata s and Clark, 2022; profile at the site, with multiple species within the prey selection Clark and Linares-Mata s, 2023). These patterns have been observed ranges of wolves and other canids (Kruuk and Turner, 1967; Pienaar, in the field by Yravedra et al. (2011), who noted how wolf- 1969; Mattioli et al., 1995; Jedrzejewski et al., 2000; Peterson and consumed carcasses in autumn exhibit a greater degree of alter- Ciucci, 2003; Hayward et al., 2006; Yravedra et al., 2011; Lagos- ation than those consumed during winter. In Africa, we noted how Abarzuza, 2013; Fourvel et al., 2018). However, the low tooth wet season carcasses (Fig. 6cef) tend to exhibit more complete limb mark frequencies and the low number of tooth marks per bone bones than those hunted by lions and scavenged by hyaenas during does not fit the general pattern observed for canids (Haynes, 1980, the dry season (Fig. 6b). Open-air sites tend to have skeletal ele- 1983; Yravedra et al., 2011; Fosse et al., 2012). Future research ments in anatomical connection, resembling the patterns described employing geometric morphometrics and machine learning 17 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 approaches (Aramendi et al., 2017a; Courtenay et al., 2019, 2021; assemblages with a known geospatial provenance sensu stricto. As Mate -Gonz alez et al., 2019; Yravedra et al., 2022b; Courtenay et al., such, the continuity of systematic and rigorous excavations at this 2023) may offer additional insights into the specific role of different relevant palaeontological site promises to expand our knowledge of carnivores in the formation and alteration of the two VM4 Early Pleistocene ecosystem dynamics in the southern Iberian assemblages. Peninsula. An aspect that we are particularly interested in is the In addition to any carnivore-generated accumulation, it is seasonality of bone accumulation. This variable has a considerable plausible that natural deaths played also an important role in the impact in the formation of open-air sites, since predation cycles and formation of the VM4-II fossil assemblage, given the increase in the carcass modification intensity are related to the spatio-temporal representation of Size 5 adult individuals (Table 3), including the distribution of water sources. Periods of low water availability, presence of a rhinoceros carcass with articulated skeletal elements such as Mediterranean summers, favour the concentration of prey without tooth marks or any other bone surface modifications, around shrinking pools and waterholes, and the corresponding suggesting that their remains were quickly buried prior to any creation of predation opportunities. Therefore, understanding the carnivore interaction. impact of seasonality on inter-assemblage variability ought to To summarize, and being cautious about the reliability and other become a future research priority at Early Pleistocene palae- issues associated with the VM3 sample, we can agree that VM3 ontological localities. could represent a P. brevirostris den site (Arribas, 1999; Arribas and Palmqvist, 1998, 1999; Palmqvist and Arribas, 2001). On the other Author contributions statement hand, we can conclude that the nearby site of VM4 is best described as a complex open-air site where a multiplicity of biological pro- JY: Conceptualization, investigation, methodology, supervision, cesses, including the consumption of herbivore carcasses by car- writing, formal analysis, original draft, review and editing. CL: Data nivores as well as some natural deaths, constituting a diverse faunal curation, investigation in Taphonomy, methodology, editing, orig- assemblage represented by at least two sublevels. inal draft, Visualization. JAS: Investigation in Taphonomy, Visuali- The spatial proximity of two palaeontological sites with zation. GLM: Writing, Review and Editing. VEG: Validation, formal different accumulation processes and taphonomic biographies is analysis, JJRA: Investigation in Taphonomy, Visualization, Edition, not unprecedented, since this pattern has also been documented at LAC: Investigation in Taphonomy, review, Validation. DHR: Inves- other Early Pleistocene palaeolandscapes, such as Olduvai Bed I tigation in Taphonomy. ASR: Investigation, Project Administration. (Ashley et al., 2010; Arra iz et al., 2017). Some sites, such as FLK, JC: Photography, Investigation. JS: Investigation and coordination in were dry season camps (Leakey, 1971; Domínguez-Rodrigo et al., Macrovertebrates Taxonomy. CSB: Investigation in Microfauna 2014; Linares-Mata s and Clark, 2022), in the vicinity of other sites Palaeontology. H-AB: Investigation in Microfauna Palaeontology. with relevant anthropogenic signatures, such as DS, PTK o AGS SV-K: Investigation in Macrovertebrates Taxonomy. DdM: Investi- (Cobo, 2020), while others, such as AMK (Aramendi et al., 2017b) or gation in Macrovertebrates Taxonomy. BA: Investigation in Mac- FLKN (Domínguez-Rodrigo et al., 2007; Bunn et al., 2010) seem to rovertebrates Taxonomy. OO: Investigation and Supervision in have constituted carnivore accumulations, whereas others likely Stratigraphy-Geology. JA: Investigation and Supervision in Micro- were kill sites near freshwater sources, such as FLKNN (Domínguez- fauna Palaeontology. MF: Investigation and supervision in Macro- Rodrigo et al., 2007; Ashley et al., 2010). This is therefore the first vertebrates Taxonomy, Review. JMJA: Formal analysis, project time that differences in the use of space can be proposed across administration, resources, supervision and writing, original draft, penecontemporaneous sites within a relatively small area of the review and editing. Orce basin (cf., Granados et al., 2021). However, further research must be carried out at the Venta Micena localities, considerably Declaration of competing interest expanding the excavation area in order to unravel the complex palaeoecological dynamics unfolding at Orce. The authors declare that they have no known competing financial interests or personal relationships that could have 5. Conclusions appeared to influence the work reported in this paper. The conclusions of this paper can be summarised as follows. Data availability 1. VM illustrates some of the complex site formation processes All data is included in the text and supplementary files. represented in the Early Pleistocene deposits of the Guadix-Baza Basin, where multiple and different taphonomic biographies are Acknowledgements recorded vertically and horizontally. Therefore, it is paramount to dissect these palimpsests in order to study specific assem- This research has been possible thanks to the support and blages independently, isolating them from other diachronic or approval of the Consejería de Cultura de la Junta de Andalucía synchronic events taking place in the same landscape. through the General Research Project Primeras ocupaciones 2. The VM4 faunal assemblage formed through the behaviour and humanas y contexto paleoecolo gico a partir de los depo sitos Plio- input of multiple agents, rather than by the selective transport pleistocenos de la cuenca Guadix-Baza. Zona arqueolo gica de la of any given taxon. cuenca de Orce (Granada, Espan~ a) (BC.03.032/17). Special thanks are 3. Based on the data gathered so far, it seems that VM4-is not a given to the Archaeological Museum of Granada staff, headed by its carnivore den site; instead, it resembles more closely an open- Director Isidro Toro Moyano, for its availability and kindness. We air site where multiple carnivores interacted with herbivore are grateful to the “Archaeometrical Studies. Inside the Artefacts & carcasses of different species, alongside some potential inputs Ecofacts” Excellence Unit (University of Granada) for its support. from natural deaths at VM4-II of large animals, such as We also thank the TIDOP Group from the Department of Carto- rhinoceros. graphic and Land Engineering of the Higher Polytechnics School of Avila, University of Salamanca. We would also like to recognize the While only a relatively small excavation surface has been technical support provided by C.A.I. Arqueometry and Archaeo- investigated at VM4 (39 m2), it is one of the largest Orce logical Analysis from the Complutense University of Madrid which 18 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 has been very useful in carrying out the present paper. We would accumulators? Paleontological and ecological implications of a modern bone assemblage from Olduvai gorge. PLoS One 11, e0153797. like to thank the whole Orce Research Project team, including the lisis y modelizacio n de la Tafonomía del yacimiento de Venta Arribas, A., 1999. Ana volunteers who year after year join to undertake fieldwork activ- Micena (Orce, Granada) y su estudio comparativo con otras localidades espa- ities. B. Azanza and D. DeMiguel belong to the Gobierno de Arago n ~ olas del Plio-Pleistoceno continental. Ph.D. Thesis. Universidad Complutense n Research Group E33_20 R. J.M. Jime nez-Arenas belongs to the Junta de Madrid. Arribas, A., Palmqvist, P., 1999. El modelo de actividad de Pachycrocuta en Venta de Andalucía Research Group HUM-607. This research has been Micena: Implicacio n de los hie nidos en el registro paleomastolo gico Plio- carried out thanks to projects PID 2021.125098NB.I00 funding by Pleistoceno espan ~ ol. Temas Geolo gico-Mineros ITGE 26, 173e181. MCIN/AEI/10.13039/501100011033/FEDER Una manera de hacer Arribas, A., Palmqvist, P., 1998. Taphonomy and palaeoecology of an assemblage of large mammals: Hyaenid activity in the Lower Pleistocene site at Venta Micena Europa and ProyExcel_00274 funding by Direccio n General de (Orce, Guadix-Baza Basin, Granada, Spain). Geobios 31, 3e47. Planificacio n de la Investigacio n (Consejería de Universidad, Inves- Arribas, A., Garrido, G., 2008. Hie nidos [Pachycrocuta brevirostris (Aymard, 1846) y tigacion e Innovacio n, Junta de Andalucía). C. Sa nchez-Bandera, H.- Hyaena brunnea Thunberg,1820] del yacimiento de Fonelas P-1 (cuenca de Guadix, Granada). In: Arribas, A. (Ed.), Vertebrados del Plioceno superior ter- A. Blain, and J. Agustí belong to the Consolidated Research Group minal en el suroeste de Europa: Fonelas P-1 y el Proyecto Fonelas. Instituto PalHum (Paleoecology of Pliocene and Pleistocene and Human Geologico y Minero de Espan ~ a, serie Cuadernos del Museo Geominero 10, Dispersals, 2021 SGR 01238) of the Generalitat de Catalunya Madrid, pp. 201e230. Arsznov, B.M., Lundrigan, B.L., Holekamp, K.E., Sakai, S.T., 2010. Sex and the frontal (AGAUR). C. Sa nchez-Bandera is supported by a FPI Predoctoral cortex: a developmental CT study in the spotted hyena. Brain Behav. Evol. 76, Scholarship (PRE2020-094482) associated to project CEX2019- 185e197. 000945-M-20-1 with the financial sponsorship of the Spanish Ashley, G.M., Barboni, D., Dominguez-Rodrigo, M., Bunn, H.T., Mabulla, A.Z.P., Diez- Martin, F., Barba, R., Baquedano, E., 2010. A spring and woodland: new paleo- Ministry of Science and Innovation. environmental reconstruction of FLK-Zinj, Olduvai Gorge, Tanzania. Quat. Int. 74, 304e314. Appendix A. Supplementary data Barsky, D., Celiberti, V., Cauche, D., Gre goire, S., Lebegue, F., de Lumley, H., Toro- Moyano, I., 2010. Raw material discernment and technological aspects of the Barranco Leo n and Fuente Nueva 3 stone assemblages (Orce southern Spain). Supplementary data to this article can be found online at Quat. Int. 223e224, 201e219. https://doi.org/10.1016/j.quascirev.2023.108031. Bearder, S., 1977. Feeding habits of spotted hyaenas in a woodland habitat. Afr. J. Ecol. 15, 263e280. Behrensmeyer, A., 1978. Taphonomic and ecologic information from bone weath- References ering. Paleobiology 4, 150e162. Binford, L.R., 1981. Bones: Ancient Men and Modern Myths. New York Academic Agustí, J., 1985. Bioestratigrafía de los depo sitos Plio-Pleistocenos de la depresio  n de Press, p. 320. Guadix-Baza. (Prov. Granada). Paleontologia i Evolucio  18, 13e18. Blumenschine, R., 1988. An experimental model of the timing of hominid and Agustí, J., Blain, H.A., Furio  , M., De Marfa, R., Santos-Cubedo, A., 2010. The early carnivore influence on archaeological bone assemblages. J. Archaeol. Sci. 15, Pleistocene small vertebrate succession from the Orce region (Guadix-Baza 483e502. Basin, SE Spain) and its bearing on the first human occupation of Europe. Quat. Blumenschine, R., 1995. Percussion marks, tooth marks, and experimental de- Int. 223e224, 162e169. terminations of the timing of hominid and carnivore access to long bones at FLK Agustí, J., Oms, O., Pare s, J.M., 2007. Biostratigraphy, paleomagnetism and geology of Zinjanthropus, Olduvai Gorge, Tanzania. J. Hum. Evol. 29, 21e51. the Orce ravine (Southern Spain). Quat. Sci. Rev. 26, 568e572. Blumenschine, R.J., 1986. Early hominid scavenging opportunities; implications of ntara García, V., Barba Egido, R., Barral del Pino, J., Crespo Ruiz, A., Eiriz Vidal, A., Alca carcass availability in the Serengeti and Ngorongoro ecosystems. BAR Interna- Falquina Aparicio, A.,  Herrero Calleja, S., Ibarra Jime nez, A., Megías Gonza lez, M., tional Series No 283, 163. Perez Gil, M., Pe rez Tello, V., Rolland Calvo, J., Yravedra Sa inz de los Terrenos, J., Borja, C., García-Pacheco, M., Olivares, E., Scheuenstuhl, G., Lowenstein, J., 1997. Vidal, A., Domínguez-Rodrigo, M., 2006. Determinacio  n de procesos de fractura Immunospecificity of albumin detected in 1.6-million-year-old fossils from sobre huesos frescos: un sistema de ana lisis de los angulos de los planos de Venta Micena in Orce, Granada, Spain. Am. J. Phys. Anthropol. 103, 433e441. fracturacio  n como discriminador de agentes bio ticos. Trab. Prehist. 63 (1), Bourguignon, L., Crochet, J., Capdevila, R., Ivorra, J., Antoine, P., Agustí, J., Barsky, D., 37e45. Blain, H., Boulbes, N., Bruxelles, L., Claude, J., Cochard, D., Filoux, A., Firmat, C., Anado n, P., De Deckker, P., Juli a, R., 1986. The Pleistocene lake deposits of the NE Lozano-Ferna ndez, I., Magniez, P., Pelletier, M., Rios-Garaizar, J., Testu, A., Baza Basin (Spain): salinity variations and ostracod succession. Hydrobiologia Valensi, P., De Weyer, L., 2016. Bois-de-Riquet (Le zignan-la-Ce be, He rault): a 143, 199e208. late Early Pleistocene archeological occurrence in southern France. Quat. Int. Anado n, P., Julia , R., De Deckker, E., Rosso, J.C., Soulie -Marsche , I., 1987. Contribucion 393, 24e40. a la Paleolimnología del Pleistoceno inferior de la cuenca de Baza (sector Orce- Brain, C., 1969. The probable role of leopards as predators of the swartkrans aus- Venta Micena). Paleontología i Evolucio  Mem. Esp. 1, 35-72. tralopithecines. S. Afr. Archaeol. Bull. 24, 170e171. Andrews, P., 2008. Cetaceans from a possible striped hyaena den site in Qatar. Brain, C.K., 1981. The Hunters or the Hunted? an Introduction to African Cave Journal of Taphonomy 6, 255e274. Taphonomy. University of Chicago Press, Chicago, p. 376. Aramendi, J., Mate -Gonza lez, M.A., Yravedra, J., Ortega, M.C., Arriaza, M.C., Brugal, J.P., Fosse, P., 2004. Carnivores et Hommes au Quaternaire en Europe de Gonz alez-Aguilera, D., Baquedano, E., Domínguez-Rodrigo, M., 2017a. l'Ouest. Rev. Paleobiol. 23 (2), 575e595. Discerning carnivore agency through the three-dimensional study of tooth pits: Brugal, J.P., Fosse, P., Guadelli, J.L., 1997. Comparative study of bone assemblages revisiting crocodile feeding behaviour at FLK- Zinj and FLK NN3 (Olduvai Gorge, made by recent and Pleistocene hyenids. In: 8th Meeting of Working Group 1 on Tanzania). Palaeogeogr. Palaeoclimatol. Palaeoecol. 488, 93e102. Bone Modification. Hot Springs: Augustana College, Archaeology Lab., Aramendi, J., Uribelarrea, D., Arriaza, M., Arra iz, H., Barboni, D., Yravedra, J., pp. 158e187. Ortega, M., Gidna, A., Mabulla, A., Baquedano, E., Domínguez-Rodrigo, M., Bunn, H.T., 1982. Meat-eating and Human Evolution: Studies on the Diet and 2017b. The paleoecology and taphonomy of AMK (Bed I, Olduvai Gorge) and its Subsistence Patterns of Plio-Pleistocene Hominins in East Africa. Ph.D. Thesis. contributions to the understanding of the “Zinj” paleolandscape. Palaeogeogr. University of California. Palaeoclimatol. Palaeoecol. 488, 35e49. Bunn, H., Mabulla, A., Domínguez-Rodrigo, M., Ashley, G., Barba, R., Diez-Martín, F., Arilla, M., Rosell, J., Blasco, R., 2019. Contributing to characterise wild predator Remer, K., Yravedra, J., Baquedano, E., 2010. Was FLK North levels 1e2 a classic behaviour: consumption pattern, spatial distribution and bone damage on “living floor” of Oldowan hominins or a taphonomically complex palimpsest ungulate carcasses consumed by red fox (Vulpes vulpes). Archaeological and dominated by large carnivore feeding behavior? Quat. Res. 74 (3), 355e362. Anthropological Sciences 11, 2271e2291 (A). Capaldo, S., Blumenschine, R., 1994. A quantitative diagnosis of notches made by Arilla, M., Rosell, J., Blasco, R., Domínguez-Rodrigo, M., Pickering, T., 2014. The hammerstone percussion and carnivore gnawing on bovid long bones. Am. “bear” essentials: actualistic research on Ursus arctos arctos in the Spanish Antiq. 59 (4), 724e748. pyrenees and its implications for paleontology and archaeology. PLoS One 9 (7), Capaldo, S.D., 1995. Inferring Hominid and Carnivore Behavior from Dual-Patterned e102457. Archaeofaunal Assemblages. Ph.D. Dissertation, Rutgers University. iz, H., Barboni, D., Ashley, G., Mabulla, A., Baquedano, E., Domínguez- Arra Carandell Baruzzi, M., 2020. The Orce Man: Controversy, Media and Politics in Rodrigo, M., 2017. The FLK zinj paleolandscape: reconstruction of a 1.84 Ma Human Origins Research. Brill, Leiden. wooded habitat in the FLK zinj-AMK-PTK-DS archaeological complex, middle Carandell Baruzzi, M., 2013. Homínidos, dudas y grandes titulares: la controversia bed I (Olduvai gorge, Tanzania). Palaeogeogr. Palaeoclimatol. Palaeoecol. 488, del Hombre de Orce en la prensa espan ~ ola (1983-2007). Dynamis 33 (2), 9e20. 365e387. Arriaza, M., Aramendi, J., Mate -Gonzalez, M., Yravedra, J., Stratford, D., 2019. Cheheb, R., Arzarello, M., Arnaud, J., Berto, C., C aceres, I., Caracausi, S., Colopi, F., Characterising leopard as taphonomic agent through the use of micro- Daffara, S., Canini, G., Huguet, R., Karambatsou, T., Sala, B., Zambaldi, M., photogrammetric reconstruction of tooth marks and pit to score ratio. Hist. Berruti, G., 2019. Human behavior and Homo-mammal interactions at the first Biol. 33 (2), 176e185. European peopling: new evidence from the Pirro Nord site (Apricena, Southern Arriaza, M., Domínguez-Rodrigo, M., Yravedra, J., Baquedano, E., 2016. Lions as bone Italy). Sci. Nat. 106 (5e6). 19 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Clark, J., Linares-Mat as, G.J., 2023. Seasonal resource categorisation and behavioral Fourvel, J., Magniez, P., Moigne, A., Testu, A., Joris, A., Lamglait, B., Vaccaro, C., adaptation among chimpanzees: implications for early hominin carnivory. Fosse, P., 2018. Wild dogs and their relatives: implication of experimental J. Anthropol. Sci. 101, 1e36. feedings in their taphonomical identification. Quaternaire 29 (1), 21e29. Cobo, L., 2020. Estudio Tafono mico y Espacial del Yacimiento Arqueolo gico DS del Garrido, G., García-Solano, J.A., Viseras, C., Soria, J.M., Arribas, A., 2010. Taphonomic Lecho I de la Garganta de Olduvai (Tanzania). Ph.D. Thesis. Universidad Com- approach to Fonelas P-1 site (late Pliocene, Guadix basin, Granada): descriptive plutense de Madrid. taphonomic characters related to hyaenid activity. In: Proc. of “1a Reunio n de Coil, R., Tappen, M., Ferring, R., Bukhsianidze, M., Nioradze, M., Lordkipanidze, D., Científicos sobre Cubiles de Hiena (y otros grandes carnívoros) en los Yacimientos 2020. Spatial patterning of the archaeological and paleontological assemblage Arqueolo gicos de la Península Ib erica”, pp. 132e146. at Dmanisi, Georgia: an analysis of site formation and carnivore-hominin Gibert, J., Palmqvist, P., 1995. Fractal analysis of the Orce skull sutures. J. Hum. Evol. interaction in Block 2. J. Hum. Evol. 143, 102773. 28, 561e575. Coumont, M., 2009. Proposition d’un re  fe rentiel taphonomique fossile de faunes Gibert, J., Agustí, A., Moy a-Sola, S., 1983. Presencia de Homo sp. en el yacimiento del issues d’avens-pie ges. Ann. Paleontol. 95 (1), 1e20. Pleistoceno inferior de Venta Micena. Paleontologia i Evolucio  mem. esp, Courtenay, L., Herranz-Rodrigo, D., Yravedra, J., V azquez-Rodríguez, J., Huguet, R., pp. 1e12. Barja, I., Mate -Gonza lez, M., Ferna ndez, M., Mun  Gonza ~ oz-Nieto, A., lez- Gibert, J., Gibert, L., Iglesias, A., Maestro, E., 1998. Two ‘Oldowan’ assemblages in the Aguilera, D., 2021. 3D insights into the effects of captivity on wolf mastication PlioPleistocene deposits of the Orce region, southeast Spain. Antiquity 72, and their tooth marks; implications in ecological studies of both the past and 17e25. https://doi.org/10.1017/S0003598X00086233. present. Animals 11 (8), 2323. Gibert, J., Gibert, L., Albadalejo, S., Ribot, F., Sanchez, F., Gibert, P., 1999. Molar tooth Courtenay, L.A., Yravedra, J., Huguet, R., Aramendi, J., Mate -Gonz alez, M.A., Gonza lez fragment BL5-0: The oldest human remain found in the Plio-Pleistocene of Orce Aguilera, D., Arriaza, M.C., 2019. Combining machine learning algorithms and (Granada province, Spain). Hum. Evol. 14, 3e19. geometric morphometrics: a study of carnivore tooth marks. Palaeogeogr. Gibert, L., Ferra ndez-Can ~ adell, C. (Eds.) 2015. Geología y Paleontología de Cueva Palaeoclimatol. Palaeoecol. 522, 28e39. Victoria. Mastia 11e13, 1e478. Courtenay, L.A., Yravedra, J., Herranz-Rodrigo, D., Rodríguez-Alba, J.J., Serrano- Gidna, A., Yravedra, J., Domínguez-Rodrigo, M., 2013. A cautionary note on the use Ramos, A., Estaca-Go  mez, V., Gonza lez-Aguilera, D., Solano, J.A., Jime nez- of captive carnivores to model wild predator behavior: a comparison of bone Arenas, J.M., 2023. Deciphering carnivoran competition for animal resources at modification patterns on long bones by captive and wild lions. J. Archaeol. Sci. the 1.46 Ma early Pleistocene site of Barranco Leo n (Orce, Granada, Spain). Quat. 40, 1903e1910. Sci. Rev. 300, 107912. https://doi.org/10.1016/j.quascirev.2022.107912. Granados, A., Oms, O., Anado n, P., Iban ~ ez-Insa, J., Kaakinen, A., Jimenez-Arenas, J., Cruz-Uribe, K., 1991. Distinguishing hyena from hominid bone accumulations. 2021. Geochemical and sedimentary constraints on the formation of the Venta J. Field Archaeol. 18, 476e486. Micena early Pleistocene site (Guadix-Baza Basin, Spain). Sci. Rep. 11. Dennell, R.W., Coard, R., Turner, A., 2008. Predators and scavengers in early Pleis- Haynes, G., 1980. Evidence of carnivore gnawing on Pleistocene and Recent tocene southern asia. Quat. Int. 192 (1), 88, 0. https://doi:10.1016/j.quaint.2007. mammalian bones. Paleobiology 6 (3), 341e351. 06.023. Haynes, G., 1981. Bone Modifications and Skeletal Disturbances by Natural D'Errico, F., Villa, P., 1997. Holes and grooves: the contribution of microscopy and Agencies: Studies in North America. Unpublished Ph.D. Thesis, Catholic Uni- taphonomy to the problem of art origins. J. Hum. Evol. 33 (1), 1e31. versity of America. Domínguez-Rodrigo, M., Barba, R., 2006. New estimates of tooth mark and per- Haynes, G., 1983. A guide for differentiating mammalian carnivore taxa responsible cussion mark frequencies at the FLK Zinj site: the carnivore-hominid-carnivore for gnaw damage to herbivore limb bones. Paleobiology 9 (2), 164e172. hypothesis falsified. J. Hum. Evol. 50 (2), 170e194. Haynes, G., 1987. Proboscidean die-offs and die-outs: age profiles in fossil collec- Domínguez-Rodrigo, M., Pickering, T., 2010. A multivariate approach for discrimi- tions. J. Archaeol. Sci. 14 (6), 659e668. nating bone accumulations created by spotted hyenas and leopards: harnessing Haynes, G., 1988. Studies of elephant death and die-offs: potential applications in actualistic data from East and Southern Africa. Journal of Taphonomy 8, understanding mammoth bone assemblages. In: Webb, E. (Ed.), Recent De- 155e179. velopments in Environmental Analysis in Old and New World Archaeology. BAR Domínguez-Rodrigo, M., Barba, R., Egeland, C.P., 2007. Deconstructing Olduvai: A International Series No. 416, Oxford, pp. 151e169. Taphonomic Study of the Bed I Sites. Springer, New York, p. 292. Haynes, G., 1991. Mammoths, Mastodonts, and Elephants: Biology, Behavior, and the Domínguez-Rodrigo, M., Bunn, H., Yravedra, J., 2014. A critical re-evaluation of bone Fossil Record. Cambridge University Press, Cambridge, p. 428. surface modification models for inferring fossil hominin and carnivore in- Hayward, M., O'Brien, J., Hofmeyr, M., Kerley, G., 2006. Prey preferences of the af- teractions through a multivariate approach: application to the FLK Zinj rican wild dog Lycaon pictus (canidae: carnivora): ecological requirements for archaeofaunal assemblage (Olduvai Gorge, Tanzania). Quat. Int. 322e323, conservation. J. Mammal. 87 (6), 1122e1131. 32e43. Hill, A., 1989. Bone modification by modern spotted Hyenas. In: Bonnichsen, R., Domínguez-Rodrigo, M., Gidna, A., Yravedra, J., Musiba, C., 2012. A comparative Sorg, M.H. (Eds.), Bone Modification. Center for the Study of the First Americans, neotaphonomic study of felids, hyenids and canids: an analogical framework Orono, pp. 169e178. based on long bone modification patterns. Journal of Taphonomy 10, 147e164. Holekamp, K.E., Sakai, S.T., Lundrigan, B.L., 2007a. Social intelligence in the spotted Domínguez-Rodrigo, M., Yravedra, J., Organista, E., Gidna, A., Fourvel, J., hyena (Crocuta crocuta). Phil. Trans. Roy. Soc. Lond. B 362, 523e538. Baquedano, E., 2015. A new methodological approach to the taphonomic study Holekamp, K.E., Sakai, S.T., Lundrigan, B.L., 2007b. The spotted hyena (Crocuta of paleontological and archaeological faunal assemblages: a preliminary case crocuta) as a model system for study of the evolution of intelligence. J. Mammal. study from Olduvai Gorge (Tanzania). J. Archaeol. Sci. 59, 35e53. 88, 545e554. Duval, M., Falgue res, C., Bahain, J., Grün, R., Shao, Q., Aubert, M., Dolo, J., Agustí, J., Holekamp, K.E., Dantzer, B., Stricker, G., Yoshida, K.C.S., Benson-Amram, S., 2015. Martínez-Navarro, B., Palmqvist, P., Toro-Moyano, I., 2012. On the limits of using Brains, brawn and sociality: a hyaena's tale. Anim. Behav. 103, 237e248. combined U-series/ESR method to date fossil teeth from two Early Pleistocene Hutson, J., 2012. Neotaphonomic measures of carnivore serial predation at Ngamo archaeological sites of the Orce area (Guadix-Baza basin, Spain). Quat. Res. 77 Pan as an analog for interpreting open-air faunal assemblages. J. Archaeol. Sci. (3), 482e491. 39 (2), 440e457. Egeland, A.G., Egeland, C., Bunn, H.T., 2008. Taphonomic analysis of a modern Hutson, J.M., Cain, C.R., 2008. Reanalysis and reinterpretation of the kalkbank faunal spotted hyena (Crocuta crocuta) den from Nairobi, Kenya. Journal of Taphonomy accumulation, limpopo province, South Africa. Journal of Taphonomy 6, 6, 275e299. 399e428. Espigares, M.P., 2010. An alisis y modelizacio  n del contexto sedimentario y los Iannucci, I., Mecozzi, B., Sardella, R., Iurino, D.A., 2021. The extinction of the giant atributos tafono micos de los yacimentos pleistoce nicos del borde nororiental hyena Pachycrocuta brevirostris and a reappraisal of the epivillafranchian and de la cuenca de Guadix-Baza. Universidad de Granada, Granada. galerian hyaenidae in Europe: faunal turnover during the earlyemiddle Pleis- Fernandez-Jalvo, Y., Andrews, P., 2016. Atlas of Taphonomic Identifications. Springer. tocene transition. Quat. Sci. Rev. 272, 107240. https://doi.org/10.1016/ https://doi.org/10.1007/978-94-017-7432-1. j.quascirev.2021.107240. Faith, J., 2007. Sources of variation in carnivore tooth-mark frequencies in a modern Jedrzejewski, W., Jedrzejewska, B., Okarma, H., Schmidt, K., Zub, K., Musiani, M., spotted hyena (Crocuta crocuta) den assemblage, Amboseli Park, Kenya. 2000. Prey selection and predation by wolves in Bialowieza primeval forest, J. Archaeol. Sci. 34, 1601e1609. Poland. J. Mammal. 81, 197e212. Fosse, P., 1994. Taphonomie pale olithique: Les grands mammife res de Soleilhac Katsagoni, A., Konidaris, G.E., Giusti, D., Harvati, K., Kostopoulos, D.S., 2022. Bone (Haute-Loire) et de Lunel-Viel 1 (He rault). Ph.D. Thesis. Universite  de Provence modifications by the giant hyaena Pachycrocuta brevirostris on large-sized un- Aix-Marseille, p. 257. gulate carcasses from the lower Pleistocene site of Tsiotra Vryssi (Mygdonia Fosse, P., Avery, G., Selva, N., Smietana, W., Okarma, H., Wajrak, A., Fourvel, J.B., basin, Greece). Hist. Biol. https://doi.org/10.1080/08912963.2022.2140424. Madelaine, S., 2011. Taphonomie compare e des os longs d'ongule  s de vore s par Krajcarz, M., Krajcarz, M., 2014. The red fox (Vulpes vulpes) as an accumulator of les grands pre dateurs modernes d'Europe et d'Afrique (C. lupus, P. brunnea). In: bones in cave-like environments. Int. J. Osteoarchaeol. 24 (4), 459e475. Brugal, J.P., Gardeisen, A., Zucker, A. (Eds.), Pr edateurs dans tous leurs  etats; Kruuk, H., Turner, M., 1967. Comparative notes on predation by lion, leopard,  evolution, biodiversit e, interactions, mythes, symboles. XXXI rencontres in- cheetah and wild dog in the Serengeti Area, Tanzania. Mammalia 31 (1), 1e27. ternationals d'arche ologie et d'histoire d'Antibes, pp. 127e156. Kruuk, H., 1972. The Spotted Hyena. A Study of Predation and Social Behavior. The Fosse, P., Selva, N., Wajrak, A., Fourvel, J.B., Madelaine, S., et al., 2012. Bone modi- University of Chicago Press, Chicago, p. 335. fication by modern wolf (Canis lupus): a taphonomic study from their natural Kuhn, B., Berger, L., Skinner, J., 2008. Examining criteria for identifying and differ- feeding places. Journal of Taphonomy 10 (3e4), 197e217. entiating fossil faunal assemblages accumulated by hyenas and hominins using Fourvel, J., Fosse, P., Avery, G., 2015. Spotted, striped or brown? Taphonomic studies extant hyenid accumulations. Int. J. Osteoarchaeol. 20 (1), 15e31. https:// at dens of extant hyaenas in eastern and southern Africa. Quat. Int. 369, 38e50. doi.org/10.1002/oa.996. 20 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Kuhn, B., Berger, L., Skinner, J., 2009. Variation in tooth mark frequencies on long Oms, O., Anado n, P., Agustí, J., Julia , R., 2011. Geology and chronology of the con- bones from the assemblages of all three extant bone-collecting hyaenids. tinental Pleistocene archeological and paleontological sites of the Orce area J. Archaeol. Sci. 36 (2), 297e307. (Baza basin, Spain). Quat. Int. 243, 33e43. https://doi.org/10.1016/ Lacruz, R.S., Maude, G., 2005. Bone accumulations at brown hyena (Parahyaena j.quaint.2011.03.048. brunnea) den sites in the Makgadikgadi Pans, Northern Botswana: taphonomic, Palmqvist, P., 1997. A critical re-evaluation of the evidence for the presence of behavioral and palaeoecological implications. Journal of Taphonomy 3, 43e53. hominids in lower Pleistocene times at Venta Micena, Southern Spain. J. Hum. Lagos-Abarzuza, L., 2013. Ecología del lobo (Canis lupus), del poni salvaje (Equus Evol. 33, 83e89. ferus atlanticus) y del ganado vacuno semiextensivo (Bos taurus) en Galicia. Palmqvist, P., Arribas, A., 2001. Taphonomic decoding of the paleobiological infor- Ph.D. Thesis. Universidad de Santiago de Compostela, Santiago de Compostela. mation locked in a lower Pleistocene assemblage of large mammals. Paleobi- Lam, Y., 1992. Variability in the behaviour of spotted hyaenas as taphonomic agents. ology 27 (3), 512e530. J. Archaeol. Sci. 19 (4), 389e406. Palmqvist, P., Espigares, M., Pe rez-Claros, J., Figueirido, B., Guerra-Merch an, A., Ros- Lam, Y., Chen, X., Pearson, O., 1999. Intertaxonomic variability in patterns of bone Montoya, S., Rodríguez-Go mez, G., García-Aguilar, J., Granados, A., Martínez- density and the differential representation of bovid, cervid, and equid elements Navarro, B., 2022. De ja  vu: a reappraisal of the taphonomy of quarry VM4 of the in the archaeological record. Am. Antiq. 64 (2), 343e362. early Pleistocene site of Venta Micena (Baza Basin, SE Spain). Sci. Rep. 12 (1). Lam, Y., Chen, X., Marean, C., Frey, C., 1998. Bone density and long bone represen- Palmqvist, P., Gro€cke, D., Arribas, A., Farin ~ a, R., 2003. Paleoecological reconstruction tation in archaeological faunas: comparing results from CT and photon densi- of a lower Pleistocene large mammal community using biogeochemical (d13C, tometry. J. Archaeol. Sci. 25 (6), 559e570. d15N, d18O, Sr:Zn) and ecomorphological approaches. Paleobiology 29 (2), Leakey, R., 1971. Further evidence of lower Pleistocene hominids from east rudolf, 205e229. north Kenya. Nature 231, 241e245. Palmqvist, P., Martínez-Navarro, B., Arribas, A., 1996. Prey selection by terrestrial Lyman, R.L., 1994. Vertebrate Taphonomy. Cambridge University Press. carnivores in a lower Pleistocene paleocommunity. Paleobiology 22 (4), Linares-Mat as, G.J., Fern andez Ruiz, N., Haber Uriarte, M., Lo pez-Martínez, M., 514e534. Walker, M.J., 2021. Hyaenas and early humans in the latest early Pleistocene of Palmqvist, P., Martínez-Navarro, B., Pe rez-Claros, J., Torregrosa, V., Figueirido, B., south-western Europe. Sci. Rep. 11, 24036. https://doi.org/10.1038/s41598-021- Jime nez-Arenas, J., Patrocinio Espigares, M., Ros-Montoya, S., De Renzi, M., 2011. 03547-7. The giant hyena Pachycrocuta brevirostris: modelling the bone-cracking Linares-Mat as, G.J., Yravedra, J., 2021. ‘We hunt to share’: social dynamics and very behavior of an extinct carnivore. Quat. Int. 243 (1), 61e79. large mammal butchery during the OldowaneAcheulean transition. World Palmqvist, P., Perez-Claros, J., Janis, C., Gro €cke, D., 2008a. Tracing the ecophysiology Archaeol. 53 (2), 224e254. https://doi.org/10.1080/00438243.2022.2030793. of ungulates and predatoreprey relationships in an early Pleistocene large Linares-Mat as, G.J., Clark, J., 2022. Seasonality and oldowan behavioral variability in mammal community. Palaeogeogr. Palaeoclimatol. Palaeoecol. 266 (1e2), east africa. J. Hum. Evol. 164, 103070. https://doi.org/10.1016/ 95e111. j.jhevol.2021.103070. Palmqvist, P., Pe rez-Claros, J., Janis, C., Figueirido, B., Torregrosa, V., Grocke, D., n, C., Yravedra, J., Courtenay, L., Saarinen, J., Blain, H., DeMiguel, D., Viranta, S., Luzo 2008b. Biogeochemical and ecomorphological inferences on prey selection and Azanza, B., Rodríguez-Alba, J., Herranz-Rodrigo, D., Serrano-Ramos, A., Solano, J., resource partitioning among mammalian carnivores in an early Pleistocene Oms, O., Agustí, J., Fortelius, M., Jime nez-Arenas, J., 2021. Taphonomic and community. Palaios 23 (11), 724e737. spatial analyses from the early Pleistocene site of Venta Micena 4 (Orce, Pereira, L.M., Owen-Smith, N., Moleo n, M., 2014. Facultative predation and scav- Guadix-Baza Basin, southern Spain). Sci. Rep. 11. enging by mammalian carnivores: seasonal, regional and intra-guild compari- Madurell-Malapeira, J., Alba, D.M., Aurell-Garrido, J., Moy a-Sola , S., 2011. Tapho- sons. Mamm Rev. 44, 44e55. https://doi.org/10.1111/mam.12005. nomic approach to the last villafranchian faunas of Europe. The layer 7 of the rez-Claros, J.A., Coca-Ortega, C., 2020. Canines and carnassials as indicators of Pe vallparadís estacio  local section (terrassa, NE iberian peninsula). In: IX sociality in durophagous hyaenids: analyzing the past to understand the pre- Encuentro de Jo  venes Investigadores en Paleontologia. Morella, Castello  n. sent. PeerJ 8, e10541. https://doi.org/10.7717/peerj.10541. Madurell-Malapeira, J., Morales, J., Vinuesa, V., Boscaini, A., 2015. Úrsidos, hie nidos Peterson, R.O., Ciucci, P., 2003. The wolf as a carnivore. In: Mech, D., Botani, L. (Eds.), y felidos del Pleistoceno inferior de Cueva Victoria (Cartagena, Murcia). In: Wolves, Behavior, Ecology and Conservation. The University of Chicago Press, Gibert, L., Fern andez Canadell, C. (Eds.), Geología y Paleontología de Cueva Chicago, pp. 104e130. Victoria. Mastia: Revista del Museo Arqueol_ogico Municipal de Cartagena. Pickering, T., 2002. Reconsideration of criteria for differentiating faunal assemblages Ayuntamiento de Cartagena, Cartagena, pp. 401e432. accumulated by hyenas and hominids. Int. J. Osteoarchaeol. 12 (2), 127e141. Madurell-Malapeira, J., Alba, D.M., Espigares, M.P., Vinuesa, V., Palmqvist, P., Mar- Pickering, T.R., Heaton, J.L., Zwodeski, S.E., Kuman, K., 2011. Taphonomy of bones tínez- Navarro, B., Moy , S., 2017. Were large carnivorans and great cli- a-Sola from baboons killed and eaten by wild leopards in Mapungubwe National Park, matic shifts limiting factors for hominin dispersals? Evidence of the activity of South Africa. J. Taphonomy 9, 117e159. Pachycrocuta brevirostris during the Mid-Pleistocene Revolution in the Vall- Pienaar, U., 1969. Predator-prey relationships amongst the larger mammals of the paradís Section (Vall_es-Pened_es Basin, Iberian Peninsula). Quat. Int. 431, Kruger National Park. Koedoe 12 (1), 108e176. 42e52. https://doi.org/10.1016/j.quaint.2015.07.040. Pobiner, B., Blumenschine, R., 2003. A taphonomic perspective on Oldowan hominid Marean, C., Kim, S., 1998. Mousterian large-mammal remains from kobeh cave encroachment on the carnivoran paleoguild. J. Taphonomy 1, 115e141. (zagros mountains, Iran): behavioral implications for neanderthals and early Pokines, J., Kerbis Peterhans, J., 2007. Spotted hyena (Crocuta crocuta) den use and modern humans. Curr. Anthropol. 39 (S1), S79eS114. taphonomy in the Masai Mara National Reserve, Kenya. J. Archaeol. Sci. 34 (11), Marra, A., Villa, P., Beauval, C., Laura, B., Goldberg, P., 2004. Same predator, variable 1914e1931. prey: taphonomy of two Upper Pleistocene hyena dens in Sicily and SW France. Prendergast, M.E., Domínguez-Rodrigo, M., 2008. Taphonomic analyses of a hyena Rev. Paleobiol. 23, 787e801. den and a natural-death assemblage near lake Eyasi (Tanzania). J. Taphonomy 6, Martínez-Navarro, B., 1991. Revisio n sistema tica y estudio cuantitativo de la fauna 301e335. de macromamíferos del yacimiento de Venta Micena (Orce, Granada). Ph.D. ndez-Can Ribot, F., Gibert, L., Ferra ~ adell, C., García-Olivares, E., Sa nchez, F., Lería, M., Thesis. Universitat Auto noma de Barcelona, Barcelona. 2015. Two deciduous human molars from the Early Pleistocene deposits of Martínez-Navarro, B., 2002. The skull of Orce: parietal bones or frontal bones? Barranco Leo  n (Orce, Spain). Curr. Anthropol. 56, 134e142. J. Hum. Evol. 42, 265e270. Rodríguez-Alba, J., Linares-Mata s, G., Yravedra, J., 2019. First assessments of the Mate-Gonz alez, M.A., Gonza lez-Aguilera, D., Linares-Mat as, G., Yravedra, J., 2019. taphonomic behaviour of jaguar (Panthera onca). Quat. Int. 517, 88e96. New technologies applied to modelling taphonomic alterations. Quat. Int. 517, Ruiter, D., Berger, L., 2000. Leopards as taphonomic agents in dolomitic cav- 4e15. https://doi.org/10.1016/j.quaint.2018.12.021. esdimplications for bone accumulations in the hominid-bearing deposits of Mattioli, L., Apollonio, M., Mazzarone, V., Centofanti, E., 1995. Wolf food habits and South Africa. J. Archaeol. Sci. 27 (8), 665e684. wild ungulate availability in the Foreste Casentinesi National Park, Italy. Acta Sakai, S.T., Arsznov, B.M., Lundrigan, B.L., Holekamp, K.E., 2011. Brain size and social Theriol. 40, 387e402. complexity: a computed tomography study in Hyaenidae. Brain Behav. Evol. 77, Mazza, P., Bertini, A., Magi, M., 2004. The late pliocene site of poggio rosso (Central 91e104. Italy): taphonomy and paleoenvironment. Palaios 19 (3), 227e248. Saunders, J., Dawson, B., 1998. Bone damage patterns produced by extinct hyena, Michel, P., 2004. La grotte dʼUnikote  (Iholdy, Pyre ne es Atlantiques, France): une Pachycrocuta brevirostris (mammalia: carnivora) at the Haro River Quarry, grotte-repaire dʼhye nes des cavernes avec des indices de pre sence humaine northwestern Pakistan. In: Yomida, Y., Flynn, L.J., Jacobs, L.L. (Eds.), Advances in dans des niveaux würmiens. Rev. Pale obiol. 23, 741e760. Vertebrate Paleontology and Geochronology, 14. National Science Museum Mills, M., 1989. The comparative behavioral ecology of hyenas: the importance of Monographs, Tokyo, pp. 215e242. diet and food dispersion. In: Gittleman, J.L. (Ed.), Carnivore Behavior, Ecology, Sept, J., 1994. Bone distribution in a semi-arid riverine habitat in eastern Zaire: and Evolution. Springer, Boston, pp. 125e142. implications for the interpretation of faunal assemblages at early archaeological Mocla n, A., Domínguez-Rodrigo, M., Yravedra, J., 2019. Classifying agency in bone sites. J. Archaeol. Sci. 21 (2), 217e235. breakage: an experimental analysis of fracture planes to differentiate between Shipman, P., 1981. Life History of a Fossil. An Introduction to Taphonomy and hominin and carnivore dynamic and static loading using machine learning (ML) Paleoecology. University Press, Harvard. algorithms. Archaeol. Anthropol. Sci. 11 (9), 4663e4680. Sinclair, A., Mduma, S., Brashares, J., 2003. Patterns of predation in a diverse Moy a-Sola, S., Ko €hler, M., 1997. The Orce skull: anatomy of a mistake. J. Hum. Evol. predator-prey system. Nature 425, 288e290. https://doi.org/10.1038/ 33, 91e97. nature01934. Mwebi, O., Brugal, J., 2018. Comparative taphonomical studies of sympatric hyenids Skinner, J., Henschel, J., van Jaarsveld, A., 1986. Bone-collecting habits of spotted (Crocuta crocuta and Hyaena hyaena) bone assemblages, insights from modern hyaenas Crocuta crocuta in the Kruger National Park. S. Afr. J. Zool. 21 (4), dens in Kenya. Quaternaire 29 (1), 13e20. 303e308. 21 n, J.A. Solano et al. J. Yravedra Sainz de los Terreros, C. Luzo Quaternary Science Reviews 308 (2023) 108031 Sutcliffe, A., 1970. Spotted Hyaena: crusher, gnawer, digester and collector of bones. den: an integrated approach. J. Archaeol. Sci. 37 (5), 919e935. Nature 227 (5263), 1110e1113. Vinuesa, V., Madurell-Malapeira, J., Fortuny, J., Alba, D.M., 2015. The endocranial Tappen, M., 1992. Taphonomy of a central African savanna: natural bone deposition morphology of the Plio-Pleistocene bone-cracking hyena Pliocrocuta perrieri: in Parc National des Virunga, Zaire. Ph.D. Thesis. Harvard University, Cambridge. behavioral implications. J. Mamm. Evol. 22, 421e434. Tappen, M., Lordkipanidze, D., Bukshianidze, M., Ferring, R., Vekua, A., 2007. Are you Vinuesa, V., Iurino, D.A., Madurell-Malapeira, J., Liu, J., Fortuny, J., Sardella, R., in or out (of Africa)? Site formation at Dmanisi and actualistic studies in Africa. Alba, D.M., 2016. Inferences of social behavior in bone-cracking hyaenids In: Pickering, T., Schick, K., Toth, N. (Eds.), Breathing Life into Fossils: Taphonomic (Carnivora, Hyaenidae) based on digital paleoneurological techniques: impli- Studies in Honor of C.K. Brain. Stone Age Institution Press, Bloomington, cations for human-carnivoran interactions in the Pleistocene. Quat. Int. 413, pp. 119e135. 7e14. https://doi.org/10.1016/j.quaint.2015.10.037. Thompson, C., Ball, S., Thompson, T., Gowland, R., 2011. The abrasion of modern and Young, A., Stillman, R., Smith, M., Korstjens, A., 2015. Scavenger species-typical archaeological bones by mobile sediments: the importance of transport modes. alteration to bone: using bite mark dimensions to identify scavengers. J. Archaeol. Sci. 38 (4), 784e793. J. Forensic Sci. 60 (6), 1426e1435. Titton, S., Barsky, D., Bargallo , A., Verge s, J.M., Guardiola, M., García-Solano, J., Yravedra, J., Domínguez-Rodrigo, M., 2009. The shaft-based methodological Jime nez-Arenas, J.M., Toro-Moyano, I., Sala-Ramos, R., 2018. Active percussion approach to the quantification of long limb bones and its relevance to under- tools from the Oldowan site of Barranco Leo n (Orce, Andalusia, Spain): The standing hominid subsistence in the Pleistocene: application to four Palae- fundamental role of punding activities in hominin lifeways. J. Archaeol. Sci. 96, olithic sites. J. Quat. Sci. 24 (1), 85e96. 131e147. https://doi.org/10.1016/j.jas.2018.06.004. Yravedra, J., Andre s, M., Domínguez-Rodrigo, M., 2013. A taphonomic study of the Titton, S., Barsky, D., Bargallo  , A., Serrano-Ramos, A., Verge s, J.M., Toro-Moyano, I., African wild dog (Lycaon pictus). Archaeological and Anthropological Sciences 6 Sala-Ramos, R., García-Solano, J., Jime nez-Arenas, J.M., 2020. Subspheroids in (2), 113e124. the lithic assemblage of Barranco Leo n (Spain): recognizing the late Oldowan in Yravedra, J., Andre s, M., Fosse, P., 2014. Taphonomic analysis of small ungulates Europe. PLoS One 15 (1), e0228290. https://doi.org/10.1371/ modified by fox (Vulpes vulpes) in southwestern Europe. J. Taphonomy 12 (1), journal.pone.0228290. 37e67. Titton, S., Oms, O., Barsky, D., Bargallo  , A., Serrano-Ramos, A., García-Solano, J., Yravedra, J., Lagos, L., Ba rcena, F., 2011. A taphonomic study of wild wolf (Canis Sanchez-Bandera, C., Yravedra, J., Blain, H., Toro-Moyano, I., Jime nez Arenas, J., lupus) modification of horse bones in northwestern Spain. J. Taphonomy 9, Sala-Ramos, R., 2021. Oldowan stone knapping and percussive activities on a 37e66. raw material reservoir deposit 1.4 million years ago at Barranco Leo n (Orce, Yravedra, J., Lagos, L., Barcena, F., 2012. The wild wolf (Canis lupus) as a dispersal Spain). Archaeol. Anthropol. Sci. 13, 108. agent of animal carcasses in northwestern Spain. J. Taphonomy 10, 227e248. Tixier, J., Roe, D., Turq, A., Gibert, J., Martínez-Navarro, B., Arribas, A., Gibert, L., Yravedra, J., Solano, J.A., Courtenay, L.A., Saarinen, J., Linares-Mat n, C., as, G., Luzo Maillo, A., Iglesias, A., 1995. Presence d’industries lithiques dans le Ple istoce ne Serrano-Ramos, A., Herranz-Rodrigo, D., C amara, J.M., Ruiz, A., Titton, S., inferieur de la re gion d'Orce (Grenade, Espagne): etat de la question. Comptes Rodríguez-Alba, J.J., Mielgo, C., Blain, H.-A., Agustí, A., Sa nchez-Bandera, C., Rendus de la Academie des Sciences de Paris se rie IIa 321, 71e78. Montilla, E., Toro-Moyano, I., Fortelius, M., Oms, O., Barsky, D., Jime nez- Toro-Moyano, I., Martínez-Navarro, B., Agustí, J., Souday, C., Bermúdez de Castro, J., Arenas, J.M., 2021. Use of meat resources in the early Pleistocene assemblages Martino n-Torres, M., Fajardo, B., Duval, M., Falgue res, C., Oms, O., Pare s, J., from Fuente Nueva 3 (Orce, Granada, Spain). Archaeological and Anthropo- Anado  n, P., Julia , R., García-Aguilar, J., Moigne, A., Espigares, M., Ros-Montoya, S., logical Sciences 13, 213. https://doi.org/10.1007/s12520-021-01461-7. Palmqvist, P., 2013. The oldest human fossil in Europe, from Orce (Spain). Yravedra, J., Solano, J.A., Herranz-Rodrigo, D., Linares-Mat as, G., Saarinen, J., J. Hum. Evol. 65 (1), 1e9. Rodríguez-Alba, J.J., Serrano-Ramos, A., Mielgo, C., Courtenay, L.A., Titton, S., Turner, A., Anto n, M., 1996. The giant hyaena Pachycrocuta brevirostris (mammalia, Blain, H.-A., Luzo n, C., Ca mara, J.M., Agustí, A., Sa nchez-Bandera, C., Montilla, E., carnivora, hyaenidae). Geobios 29, 455e468. https://doi.org/10.1016/S0016- Toro-Moyano, I., Fortelius, M., Oms, O., Barsky, D., Jime nez-Arenas, J.M., 2022a. 6995(96)80005-2. Unravelling hominin activities in the zooarchaeological assemblage of Barranco Turq, A., Martínez-Navarro, B., Palmqvist, P., Arribas, A., Agustí, J., Rodríguez- Leon (Orce, Granada, Spain). J. Palaeolithic Archaeol. 5, 6. https://doi.org/ Vidal, J., 1996. Le Plio-Ple istoce ne de la re gion d'Orce, province de Grenade, 10.1007/s41982-022-00111-1. Espagne: bilan et perspectives de recherche. Pale o 8, 161e204. Yravedra, J., Courtenay, L.A., Herranz-Rodrigo, D., Rodríguez-Alba, J.J., Linares- Villa, P., Mahieu, E., 1991. Breakage patterns of human long bones. J. Hum. Evol. 21 Mat as, G., Serrano-Ramos, A., Solano, J.A., Jime nez-Arenas, J.M., 2022b. The first (1), 27e48. morphological characterization of an extinct carnivore's tooth marks from the nchez Gon Villa, P., Sa ~ i, M., Bescos, G., Grün, R., Ajas, A., García Pimienta, J., Lees, W., Eurasian Early Pleistocene. Sci. Bull. https://doi.org/10.1016/j.scib.2022.07.017. 2010. The archaeology and paleoenvironment of an Upper Pleistocene hyena 22