Bollettino della Società Paleontologica Italiana, 51 (2), 2012, 137-148. Modena, 28 settembre 2012

ISSN 0375-7633 doi:10.4435/BSPI.2012.15

of fossils previously attributed to the genus Cuon [i.e., C. dubius Teilhard de Chardin, 1940 and C. dubius stehlini Thenius, 1954 (= C. stehlini Bonifay, 1971)], were recently reconsidered and different forms have also been attributed to Xenocyon (Sotnikova, 2001; Moullé et al., 2006; Echassoux et al., 2008) or Lycaon (Martinez-Navarro & Rook, 2003), largely thanks to the discovery and description of new fossil material. Therefore the main diagnostic features of the genus Cuon are the middle-sized body, a basically “blade-like” talonid in the lower carnassial tooth and, above all, the absence of the third lower molar (M3).

The earliest fossil species referable to Cuon recognized in Europe was C. priscus described by Thenius (1954), on the basis of fossil remains from the Middle Pleistocene site of Hundsheim (Austria), including a mandible lacking the M3. This species is poorly represented, but has been recorded in several Middle Pleistocene European localities such as Mosbach (Germany), Lunel-Viel and Caune de l’Arago (France), and Galeria Pesada (Portugal) (respectively, Thenius, 1954; Bonifay, 1971; Trinkaus et al., 2003; Moigne et al., 2006). The occurrence of C. priscus in the Early Pleistocene Spanish site of Venta Micena, reported by Pons Moyà (1987), has recently been rejected after direct revision of the fossil material by

INTRODUCTION

The extant Asiatic wild dog Cuon alpinus (Pallas, 1811), also known as dhole, is a hypercarnivorous middle-sized social canid, today living in Southeastern Asia. It includes several subspecies, although not all the authors agree on their status and number (Durbin et al., 2004; Iyengar et al., 2005). The habitats of the dhole are characterized by the availability of water, and suitability of breeding sites, in which medium to large ungulates prey species are abundant. Dense forest and thick scrub jungle are preferred; deserts are avoided. The dholes are social hunters, generally forming packs of around ten individuals, occasionally up to thirty, but they are equally able to hunt alone or in pairs, depending on the size and availability of preys, which consist mainly of chital, sambar, wild sheep, butalso cattle and lagomorphs (Durbin et al., 2004).

During the Pleistocene, the dhole spread from Central Asia to whole Eurasia, reaching Central and Western Europe at the beginning of the Middle Pleistocene and North America in the Late Pleistocene (Tedford et al., 2009).

The evolutionary history of this genus is still a matter of debate and its origins remain unclear. A large number

The Middle-Late Pleistocene Cuon Hodgson, 1838 (Carnivora, Canidae) from Italy

Mauro Petrucci, Serena romiti & Raffaele Sardella

M. Petrucci, IsIPU, Istituto Italiano di Paleontologia Umana, via Aldrovandi 18, I-00197 Roma, Italy; mauro.petrucci76@gmail.comS. Romiti, via Calzatora 57, I-03012, Anagni, Italy; sere.romiti@yahoo.itR. Sardella, Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185, Roma, Italy; IsIPU, Istituto Italiano di

Paleontologia Umana, via Aldrovandi 18, I-00197 Roma, Italy; raffaele.sardella@uniroma1.it

KEY WORDS - Canidae, hypercarnivory, taxonomy, Middle-Late Pleistocene, Italian Peninsula.

ABSTRACT - The aim of this work is to review the Middle-Late Pleistocene fossil remains of the genus Cuon Hodgson, 1838 from Italy. The geographical distribution of hypercarnivorous canids belonging to the genus Cuon is presently restricted to Southeastern Asia, whereas during the Pleistocene occurrences are documented in Eurasia and North America. Due to the fragmentary nature of the fossil record of this genus, resolution of many aspects of its origins, phyletic relationships, and evolutionary trends prove difficult.

In the Italian Peninsula the occurrence of this canid has been reported from several sites of Middle-Late Pleistocene age. However, the latter have rarely been accompanied by descriptions of the material. This paper presents an update of the Italian record of the genus Cuon, description of the material from these different sites and includes unpublished data related to new localities. It represents a starting point for a broader review of the Mediterranean Plio-Pleistocene hypercarnivorous canids.

RIASSUNTO - [Cuon Hodgson, 1838 nel Pleistocene Medio e Superiore italiano] - Le specie appartenenti alla famiglia dei canidi sono rappresentate in modo relativamente frequente nei depositi pleistocenici europei. Nell’ambito di questa famiglia, i resti fossili attribuiti al genere Cuon Hodgson, 1838 sono relativamente rari e frammentati, e restano ancora aperte molte problematiche concernenti sia la loro sistematica, sia la loro filogenesi. La principale caratteristica del genere Cuon è rappresentata dall’alto grado di specializzazione verso una dieta prevalentemente a base di carne (ipercarnivoria) che è ben evidente in particolare nei caratteri della dentatura molto specializzata. Gli studi condotti in passato da diversi autori si concentrano quindi in prevalenza sui caratteri craniodentari di Cuon e il confronto con i dati relativi ad altri canidi ipercarnivori come Xenocyon e Lycaon ha condotto gli studiosi a differenti interpretazioni.

Attualmente, l’areale del genere Cuon è localizzato nel Sud-Est Asiatico, sebbene durante il Pleistocene fosse più ampio ed esteso all’Eurasia e al Nord America, raggiungendo anche le regioni più meridionali della penisola italiana. Sebbene la presenza di questo canide sia stata riportata in differenti liste faunistiche di siti del Pleistocene Medio e Superiore, soltanto per alcuni casi è disponibile una descrizione del materiale fossile. In questo lavoro sono stati revisionati i fossili riferiti a Cuon provenienti da siti pleistocenici della penisola italiana e descritto materiale inedito. La distribuzione biocronologica di questo canide in Italia è compresa tra il tardo Pleistocene Medio e l’inizio dell’Olocene.

Bollettino della Società Paleontologica Italiana, 51 (2), 2012138

However, as noted by Mivart (1890) and recently reported by Durbin et al. (2004), the M2 is variable both in shape and dimensions. In fact, on the basis of body size and morphology of the upper and lower second molars, Mivart (1890) further distinguished two species, the northern Cuon alpinus and the southern Cuon javanicus, the latter with larger second molars.

The upper carnassial tooth is sharp and highly trenchant. The upper first molar shows the occurrence of a single talon basin between the protocone, metacone and paracone, which houses the M1 hypoconid. In the cranium, the snout is foreshortened, related to the reduction of the dental formula and the powerful bite force exerted (Holliday & Steppan, 2004).

These hypercarnivorous craniodental features have occurred in several different taxa within the tribe Canini during the Middle and Late Pleistocene but are also present in the modern day forms, thus giving rise to ambiguities in taxonomical interpretations. On account of these similarities, Simpson (1945) had included all the hypercarnivorous canids in the subfamily Simocyoninae. In fact, the dholes share these hypercarnivorous features with the extant African wild dog (Lycaon pictus) and South American bush dog (Speothos venaticus) (Durbin et al., 2004), and the extinct Pleistocene taxa Xenocyon/Lycaon (Rook, 1994; Sotnikova, 2001; Martínez-Navarro & Rook, 2003) and Cynotherium sardous (Malatesta, 1962; Abbazi et al., 2005; Lyras et al., 2006).

This taxonomic assignment was questioned by several authors (e.g., Thenius, 1954; Wozencraft, 1989; Girman et al., 1993) and is no longer accepted, particularly since molecular studies supported the inconsistence of this subfamily and the separation of the modern species. Phyletic relationships between these taxa, however, remain unclear. Some authors suggest an Asiatic origin for Xenocyon (Moullé et al., 2006; Tedford et al., 2009), while in the gradualistic evolutionary scheme suggested by Martinez-Navarro & Rook (2003), the Lycaon lineage does not include the dhole. Most probably, these genera developed similar hypercarnivorous features independently, regardless of their phyletic relationships, because of their highly predatory diet (e.g., Thenius, 1954).

MATERIALS AND METHODS

In this study, the morphological and morphometrical analyses have been limited to crania, mandibles and isolated teeth (Tabs 1 and 2), since postcranial bones are both very rare and not always properly identified.

Measurements were made by digital calliper following von den Driesch (1976). Data analysis and statistical graphic outputs were performed using SPSS software.

The material is stored in the following institutions: Museo di Storia Naturale of Florence University - Geology and Paleontology Section (MSN-IGF), Museo Paleontologico of Terni (MPT), Istituto Italiano di Paleontologia Umana in Rome (IsIPU), Earth Sciences Department of “La Sapienza” University in Rome (DSTRS), and “Museo Civico di Paleontologia e Paleoetnologia Decio De Lorentis” of Maglie (Lecce, Puglia) (MPM).

Morphological and morphometrical comparisons were made with Cuon fossils and extant and extinct canids of

Martínez-Navarro (1992), which has subsequently been reassigned to Lycaon lycaonoides.

A primitive form of Cuon alpinus characterized different Middle Pleistocene sites of Central Europe. Nehring (1890, 1891) described several fossil remains and referred them to the subspecies C. a. fossilis.

The late Middle-Late Pleistocene is characterized by the subspecies C. alpinus europaeus, described for the first time in the Mars cave (Vence, France) by Bourguignat (1875) on the basis of the occurrence of a small cusp in front of the P4 protocone, which is absent in the extant dhole. Again, fossil remains of this species are not abundant but they have a wide distribution, occurring in several Middle-Late Pleistocene Italian (e.g., Del Campana, 1923; Fabiani, 1923; Boule & De Villeneuve, 1927) and wider Eurasian sites (Cordy, 1983; Baryshnikov, 1996; Pérez Ripoll et al., 2010; Brugal & Boudadi-Maligne, 2011 inter alios), and confirming the wide geographical spread of the dhole.

The systematics and palaeogeographical trajectories of these (sub)species have changed over the years. Adam (1959) considered three successive subspecies of Cuon alpinus: C. a. priscus - C. a. fossilis - C. a. europaeus, in a gradualistic evolutionary scheme. In contrast, Bonifay (1971) suggested the occurrence of three distinct species, C. stehlini, C. priscus, and C. alpinus, which then diversified into two subspecies, C. a. fossilis and C. a. europaeus, reflecting three waves of immigration into Europe from Asia at different times in the Pleistocene.

In a recent work, Brugal & Boudadi-Maligne (2011) have hypothesized the occurrence of two dispersal events of the genus Cuon in the European continent. The first one occurred during the Middle Pleistocene and involved the species C. priscus, with the succession of two subspecies, the older C. p. priscus and the younger C. p. fossilis correspondig to C. alpinus fossilis. The second dispersal event occurred in the Late Pleistocene and involved populations of Cuon alpinus, with the subspecies C. a. europaeus.

In addition the occurrence of C. a. caucasicus was proposed for Late Pleistocene deposits in the Caucasus (Baryshnikov, 1996), and of C. a. antiquus from the late Middle Pleistocene of China (Colbert & Hooijer, 1953).

HYPERCARNIVORY IN CUON

The dhole shows the typical craniodental features of a hypercarnivorous mammal as summarized by Holliday & Steppan (2004). The dental formula is I3/3, C1/1, P4/4, M2/2 (40 teeth) and the dentition is characterized by the absence of the lower third molar. The lower carnassial tooth shows highly trenchant features, including the extreme reduction (or even absence) of the metaconid and the “blade-like” talonid (Van Valkenburg, 1991). In the talonid, the extreme reduction or loss of the entoconid occurs, associated with the development, in the central position of the talonid basin, of a conical hypoconid. Also, the lower second molar shows the occurrence of sharp cusps and their reduction in number. The protoconid and the hypoconid are aligned, whereas the metaconid is highly reduced or even absent. The posterior and anterior roots tend to coalesce, as do the respective alveoli.

139M. Petrucci et alii - Italian Middle-Late Pleistocene Cuon

Pleistocene age belonging to the genera Canis, Lycaon, Speothos and Xenocyon.

For comparison with other European dhole material, additional data were compiled from published sources and from palaeontological and osteological material in various European Institutions (Tab. 3): “Museo de Prehistoria y Paleontología Josep Gibert” of Orce (Granada, Spain); “Musée de Préhistoire Régionale de Menton” (France); “Senckenberg Forschungstation für Quartärpaläontologie” of Weimar (Germany); “Naturhistorisches Museum” of Wien (Austria); Institute of Palaeontology, University of Wien; “Museo Civico di Zoologia” of Rome (Italy).

ITALIAN FOSSILIFEROUS SITES

In the Italian Peninsula Cuon remains have been reported from eight Middle-Late Pleistocene localities (Fig. 1).

TuscanySeveral dhole remains have been found at Buca del

Tasso (Fabiani, 1923) and Caverna e Tecchia di Equi, two Late Pleistocene Mousterian sites located in the Apuan Alps (Del Campana, 1923, 1954).

Buca del Tasso, near Metato (Camaiore, Lucca), which has also yielded a femur of juvenile Homo neanderthalensis (Cotrozzi et al., 1985; Manzi, 2004), has produced a left hemi-mandible of Cuon with P2 to M1 in situ (IIPU, Pl. 1, figs 7a-b). The horizontal ramus lacks the anterior dental area and a large portion of the

DSTRS MPM

Variables SS SS2004/1

MLPMLP-1

Left maxilla M1

I3L 9.00I3W 6.00CL 11.70CW 7.50P2L 9.80P2W 5.70P3L 12.15P3W 6.30P4L 21.70P4W 10.85

P4WBL 8.20P4LPr 4.00P4LPa 11.60P4LMe 9.20

M1L 14.10 13.95M1W 14.09

C_P4L 65.80

Tab. 1 - Measurements of upper teeth of Italian dhole considered in this study. Localities: SS = S. Sidero; MLP = Melpignano. Anatomic elements: C = Canine; P = Premolar; M = Molar. Measurements: L, W = Lenght and Width of indicated tooth; P4WBL = Carnassial Blade Width of Upper Fourth Premolar; P4LPr = Protocone Length of Upper Fourth Premolar; P4LPa = Paracone Length of Upper Fourth Premolar; P4LMe = Metacone Length of Upper Fourth Premolar; C_ P4L = Length of maxillary tooth row from canine to fourth premolar). See text for explanation of Institution acronyms. All measurements are in millimetres.

DSTRS MPM MPT MPT MSN-IGF MSN-IGF IIPU

Variables SS SS 2004/2

MLPMLP-02

VS VS-01

VS VS-02

CTEIGF 7380V

CTEIGF 7381V

BT BT-01

Left mandible Right M1 Right mandible Right mandible Left M1 Right M2 Left mandible

P1L 5.40P1W 3.70P2L 9.80 8.20 10.25P2W 5.80 4.75 5.75P3L 11.50 10.25 12.00P3B 6.00 5.10 6.35P4L 13.90 15.30P4W 6.60 7.40L M1 23.31 20.90 20.75 21.79 23.80B M1 9.91 8.35 8.00 8.80 10.05M1TrL 5.94 13.70 13.60 13.89 16.80M1TrW 7.70 7.40 5.97 9.80M1PaL 5.94 5.00 5.20 5.62 6.30M1PrL 9.25 7.60 8.20 9.53 9.15M1TaL 7.05 6.00 6.00 6.17 5.65M1TaW 8.19 7.00 7.40 6.97 6.25

M2L 9.00 8.90 8.12M2W 7.00 6.90 6.34Ss M1 11.20 13.25Hs M1 24.90 26.85

P2 _ M2L 61.10

Tab. 2 - Measurements of lower teeth and mandible of Italian dholes considered in this study. Localities: SS = S. Sidero; MLP = Melpignano; VS = Valserra; CTE = Caverna e Tecchia d’Equi; BT = Buca del Tasso. Anatomic elements: P =Premolar; M = Molar. Measurements: L,W = Length and Width of indicated tooth; M1TrL = Trigonid Length of lower first molar; M1TrW = Trigonid Width of lower first molar; M1PaL = Paraconid Length of lower first molar; M1PrL = Protoconid Length of lower first molar; M1TaL = Talonid Length of lower first molar; M1TaW = Talonid Width of lower first molar; Ss M1 = Mandible Width at first molar; Hs M1 = Mandible Height at first molar; P2_M2L = Length of mandible tooth row from second premolar to second molar). See text for explanation of Institution acronyms. Measurements are in millimetres.

Bollettino della Società Paleontologica Italiana, 51 (2), 2012140

holding the second molar is present. The socket is oval, narrow and bisected, indicating coalescence of M2 roots.

Further dhole remains come from the Late Pleistocene Caverna e Tecchia di Equi (Equi Terme, Fivizzano, Massa Carrara province); the fossils were described and figured by Del Campana (1954) and are stored in the MSN-IGF. The fossil dhole sample includes second, third and fifth right metacarpals (resp. MC II, MC III, MC V), a MC III, two MC IV, and a MC V left, a left lower M1 (Pl. 1, figs 5a-b; IGF 7380V), and a posterior portion of a right mandible with M2 (Pl. 1, figs 6a-b; IGF 7381V). The lower carnassial tooth IGF 7380V is represented only by the crown, with a sharp paraconid and protoconid.

vertical ramus is missing. The mandible is stout and the two mental foramina are positioned between P2 and P3. The teeth are sharp and lie close to each other. The second premolar is short and bears only the protoconid. P3 and P4 are morphologically similar, with P4 larger than P3 and with the cusplet behind the protoconid stronger and better developed. A pronounced cingulum surrounding the posterior half of all premolars is very apparent in the fourth premolar. The carnassial tooth is highly trenchant, with the paraconid and protoconid still sharp, with a reduced metaconid and a short, rounded talonid with a clearly defined edge. The talonid basin is totally occupied by the conical hypoconid. Behind the M1, only a single alveolus

Species/Subspecies Locality Age References

Cuon alpinus Cueva de Obarreta (ES) Late Pleistocene Altuna, 1983Cuon alpinus Parpallo' (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus Durangueasado (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus La Riera (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus Gabasa (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus B. Zafarraya (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus Amalda VII (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon alpinus Trinchera Galeria (ES) Middle Pleistocene García & Arsuaga, 1997Cuon alpinus Bolinkoba (ES) Late Pleistocene Pérez-Ripoll et al., 2010

C. a. caucasicus Kudaro 3 Cave Late Pleistocene Baryshnikov, 1996C. a. europaeus Grotte Observatoire (FR) Late Pleistocene Boule & de Villeneuve, 1927C. a. europaeus Mars (FR) Late Pleistocene Bourguignat, 1875C. a. europaeus Grotte de Cotencher (FR) Late Pleistocene Dubois & Stehlin, 1933C. a. europaeus Ofenbergerhöhle (DE) Late Pleistocene Teppner, 1914C. a. europaeus Repolust-Höhle (AT) Late Pleistocene Mottl, 1951C. a. europaeus Certova dira (CZ) Late Pleistocene Nehring, 1891C. a. europaeus Sipka (CZ) Late Pleistocene Nehring, 1891C. a. europaeus Chokier (BE) Late Pleistocene Cordy, 1983C. a. europaeus Subalyuk (HU) Late Pleistocene Mottl, 1951

C. a. fossilis Heppenloch (DE) Middle Pleistocene Adam, 1959C. alpinus ssp. Matuzka (RU) Late Pleistocene Baryshnikov, 1996

Cuon cf. C. alpinus Cova Negra (ES) Late Pleistocene Pérez Ripoll et al., 2010Cuon priscus Hundsheim (AT) Middle Pleistocene Thenius, 1954Cuon priscus Mosbach (DE) Middle Pleistocene Tobien, 1957

Cuon cf. C. priscus Mosbach (DE) Middle Pleistocene Schütt, 1973Cuon dubius stehlini Rosiers (FR) Middle Pleistocene De Grossouvre & Stehlin, 1912Cynotherium sardous MonteTuttavista (IT) Late Pleistocene Abbazzi et al., 2005

Lycaon pictus MCZR (IT) extant MCZROsMam2011.0377-this paperXenocyon falconeri Upper Valdarno (I) Early Pleistocene Rook, 1994Xenocyon africanus Kroomdrai A (ZA) Early Pleistocene Rook, 1994Xenocyon africanus Olduvai II (TZ) Early Pleistocene Rook, 1994

Xenocyon antoni Fan Tsun (CN) Early Pleistocene Rook, 1994Xenocyon antoni Lok. 33 (CN) Early Pleistocene Rook, 1994Xenocyon antoni Campbellpore (PK) Early Pleistocene Rook, 1994Xenocyon dubius Zhoukoudien, loc. 18 (CN) Early Pleistocene Teilhard De Chardin, 1940Xenocyon dubius Yunxian (CN) Middle Pleistocene Echassoux et al., 2008

Xenocyon lycaonoides Cueva Victoria (ES) Early Pleistocene Pons Moyà & Moyà Solà, 1978Xenocyon lycaonoides Untermassfeld (DE) late Early Pleistocene IQW 1985/20555-this paperXenocyon lycaonoides Untermassfeld (DE) late Early Pleistocene IQW 1985/21000-this paperXenocyon lycaonoides Vallonet (FR) late Early Pleistocene Moullé et al., 2006Xenocyon lycaonoides Venta Micena (ES) late Early Pleistocene VM2255-thi paperXenocyon lycaonoides Venta Micena (ES) late Early Pleistocene VM2256-this paperXenocyon lycaonoides Venta Micena (ES) late Early Pleistocene VM7000b-this paperXenocyon lycaonoides Pirro Nord (I) late Early Pleistocene Rook, 1994Xenocyon lycaonoides Pirro Nord (I) late Early Pleistocene PN 35 (DSTSR)-this paper

Tab. 3 - List of fossil canid remains used in this study.

141M. Petrucci et alii - Italian Middle-Late Pleistocene Cuon

figs 3a-b). Preliminary observations on the fossil faunal assemblage approximate their age to the beginning of the Late Pleistocene (Squazzini, pers. comm.), although further investigations are needed to clarify the chronology of the site.

The right mandible is missing the area of the anterior teeth and the vertical ramus. Two mental foramina on the labial side are positioned respectively at the level of the P2 and P3 posterior roots. The P2, P3 and P4 are sharp and lie close to each other, almost overlapping on the lingual side. The second premolar is short and bears only the protoconid. In the third premolar, two aligned cusplets are present behind the protoconid. The P4 is larger than the P3 and a small cusplet is located anterior to the protoconid, while the cusplets behind the protoconid are better developed.

The carnassial tooth overlaps the P4 on the lingual side. The protoconid and paraconid are both sharp, whereas the metaconid is very small. In the unicuspidate talonid, the hypoconid is conical and centrally positioned; no trace of entoconid can be noticed.

The second molar is large and oval-shaped. In the occlusal surface, on the anterior side, a developed protoconid is present, with a smaller hypoconid to the posterior, and on the lingual side, a slightly pronounced metaconid is present. The cusps surround a basin and the hypoconid terminates at the posterior ridge with a crest.

All teeth are worn but particularly the occlusal surfaces of the protoconids, and P4 and M1 more than the other teeth. The M1 protoconid is longitudinally broken.

The left hemi-mandible lacks the underside and is longitudinally broken, missing all the teeth except the M1. All the alveoli are nevertheless visible and the one for the second molar reveals the presence of two sockets, which would have held a two-rooted M2. The carnassial tooth shows the same features of the right hemi-mandible M1, but it is unworn. The protoconid is triangular-shaped and quite higher than paraconid. Both are sharp. The metaconid is very small and detached from the protoconid. Adjacent to the M2 alveolus, the coronoid process rises abruptly and the condyle is short and strong, as is the angulus mandibulae.

LatiumThe occurrence of Cuon has been reported from two

late Middle Pleistocene sites: Fontana Ranuccio, in the Anagni Basin, and Colle Avarone, near the locality of Ceprano (Frosinone). For the Fontana Ranuccio faunal assemblage, a K-Ar age of 0.458 ± 0.006 Ma has been provided (Biddittu et al., 1979; Biddittu & Segre, 1984; Segre & Ascenzi, 1984). The dhole is represented here only by a partial metacarpal (Segre-Naldini, pers. comm.).

The Colle Avarone faunal assemblage could be considered slightly younger than the Fontana Ranuccio one and could possibly be referred to a cold phase of the Middle Pleistocene (perhaps MIS 10) (Biddittu & Segre, 1984), but at present, the occurrence of Cuon cannot be confirmed, since specimens in the Pofi Museum could not be verified during this re-evaluation.

ApuliaIn Southern Italy, the dhole has been recorded in the

localities of San Sidero and Melpignano (Maglie, Lecce,

Longitudinal crests surround the occlusal surfaces. The crest behind the protoconid is bifurcated (one branch angled toward the metaconid, the other one toward the hypoconid). The metaconid is highly reduced. The talonid is typically unicuspid and is surrounded by a raised edge; a small denticle is present on the lingual side, just behind the metaconid. The hypoconid is centrally positioned and conical shaped. A substantial cingulum surrounds the tooth.

The mandible fragment IGF 7381V includes the proximal part of the alveolar process with M2 and the partial anterior edge of the coronoid process. Before the second molar the posterior alveolus of M1 is present. The M2 is rounded and bears two cusps, respectively a well-developed conical protoconid and, posteriorly, a very smaller crest-like hypoconid. They are aligned on the labial side and surround an internal basin. The edge is raised and a cingulum surrounds the tooth. The roots have coalesced and are divided by a groove.

Finally, the last occurrence of C. alpinus in Italy has been documented from layer 5 of Riparo Fredian (Lucca). This site has been radiocarbon dated to 10,870 ± 119 B.P. (Cilli et al., 1998). The Cuon remains comprise a canine, a scapholunar and a rib (Dott. Cilli, pers. comm.) kept in the Museo di Archeologia-Dipartimento di Scienze Archeologiche, University of Pisa.

UmbriaIn Umbria, Cuon remains have been found from the

unpublished site of Valserra (Terni) in a karst cave and are represented by several cranial and postcranial bones stored at MPT. The cranial remains include the partial horizontal ramus of a right mandible with P2 to M2 (Pl. 1, figs 4a-b), and a left mandible, lacking the base of the horizontal ramus, and bearing only the M1 (Pl. 1,

Fig. 1 - Italian localities where Cuon remains have been collected.

Bollettino della Società Paleontologica Italiana, 51 (2), 2012142

a single rounded basin. In the lower carnassial tooth, the paraconid and protoconid are sharp and are coupled by a reduced metaconid, while in the talonid, only the central conical hypoconid is present.

RESULTS AND DISCUSSION

The morphological and morphometrical analyses of the Italian Cuon fossil remains allow comparison with other European fossils (Figs 2 and 3).

The craniodental elements (skull and upper teeth) are only represented in Italy in the Apulian sample (San Sidero and Melpignano).

The upper carnassial tooth from San Sidero has been compared to many P4s of European fossil dholes and to other fossil and extant hypercarnivorous canids. Despite the pathology, similarities with Cuon, Lycaon and Xenocyon can be observed: the P4 is narrower than in Canis and shows more trenchant features. Its dimensions fall within the range of the fossil European dholes, very close to C. alpinus from La Riera (Asturias, Spain) and to Cuon cf. C. priscus from Mosbach (Fig. 2.a).

The upper first molars from San Sidero and Melpignano reveal a sharp paracone and protocone, as noted in other European fossil Cuon, as well as in all hypercarnivorous canids. The M1 from Melpignano is triangular-shaped and its talon basin is rounded as is the case for other dhole upper first molars. Although it falls within the range of the European fossil dholes, the M1 width is smaller than in others, close to the C. alpinus fossil from Bolinkoba (Vizkaya, Spain) and overlaps the one from Trinchera Galeria (Burgos, Spain) (Fig. 2.b).

In the Italian sample, mandibles and lower teeth are better represented. The mandibles from San Sidero, Valserra and Buca del Tasso are generally stout, with the horizontal rami higher than in extant dholes. They show similarities with the dhole from Grotte de l’Observatoire of Monaco and C. a. europaeus described by Boule and De Villeneuve (1927). Mental foramina are generally located under the P2 and P3, although in the mandible from San Sidero, they are slightly more anteriorly positioned. García & Arsuaga (1997) indicate that the position of the foramina is a distinctive character for Cuon and Canis, the

Apulia). The remains, stored respectively in DSTRS and MPM, come from karstic fissures called “ventarole”, filled by reddish residual sediments and containing the fossil bones of many different vertebrate species of Late Pleistocene age (De Giuli, 1983; Bologna et al., 1994; Cipullo et al., 2006).

At San Sidero, an articulated left premaxillary and maxillary, and left hemi-mandible of a dhole stored at DSTRS have been verified. These fossils belong to the same individual and show evidence of pathologies (Pl. 1, fig. 1). Premaxillary and maxillary bones bear respectively the third incisor, and the canine, second, third and fourth premolars, and first molar. The I3 and the canine are heavily worn. The P¹ was lost during the animal lifetime and bone resorption occurred. The P2 and P3 are well preserved and unworn. The cusps of the upper carnassial tooth are unworn but the anterior root is broken. A supplementary central thin root can be observed. In M1 the talon basin crown is broken because of a traumatic event during the life of the animal, which also affected the P4 and the maxillary bone.

The horizontal ramus of the hemi-mandible bears the first, second and third premolars and the second molar, lacking the area of the anterior teeth and the coronoid process. The teeth are well preserved, unworn and sharp, except for the broken P4 and M1. The former lacks the posterior half of the crown and the associated root, while in the latter, only the talonid area and the root are present. As seen in the maxilla, trauma involving the horizontal ramus resulted in consequent deformity. In the second, well-rounded molar, the protoconid is well developed and has a crest, while the hypoconid, more posteriorly aligned, is smaller. The metaconid is only slightly developed and thus has the appearance of a denticle on the internal edge.

Despite the evident trauma and subsequent pathology in the specimen from San Sidero, it is possible to recognize the typical features of the dhole, effectively summarized by Pérez Ripoll et al. (2010) as: i) the shortening of the snout, ii) the more trenchant morphology of the teeth and iii) the absence of the third molar.

From the Apulian site of Melpignano, the dhole is represented by the upper and the lower right first molars of a single individual (Pl. 1, fig. 2). The M1 talon is reduced and a sharp paracone, metacone and protocone surround

EXPLANATION OF PLATE 1

Cranial and mandible Cuon alpinus remains of Italy.

Fig. 1 - Left premaxilla, maxilla (SS2004/1) and mandible (VS2004/2) from San Sidero, overlying a skull drawing of extant Cuon alpinus from Kurten (1968).

Fig. 2 - Right upper and lower first molars (resp. MLP-01 and MLP-02) from Melpignano.

Fig. 3 - Right mandible from Valserra (VS-01); a) occlusal view of alveolar process, b) buccal view.

Fig. 4 - Right mandible from Valserra (VS-02); a) occlusal view, b) buccal view.

Fig. 5 - Left lower first molar from Caverna e Tecchia di Equi IGF 7380V. a) buccal view, b) lingual view, c) occlusal view.

Fig. 6 - Portion of right mandible with second molar from Caverna e Tecchia di Equi IGF 7381V. a) occlusal view, b) lingual view.

Fig. 7 - Horizontal ramus of left mandible from Buca del Tasso (BT-01); a) occlusal view, b) buccal view.

Scale bars are 1 cm.

143M. Petrucci et alii - Italian Middle-Late Pleistocene Cuon Pl. 1

Bollettino della Società Paleontologica Italiana, 51 (2), 2012144

The Valserra specimens are characterized by the presence of a light cingulum on the M1, whereas the cingulum is more developed in the Buca del Tasso specimen and even stronger in the Equi and Melpignano ones, especially on the labial side.

Scattergrams for the length and width of lower carnassial tooth (Fig 3.b) show continuity in the sample of Pleistocene Cuon, with Melpignano and Equi on the top of the range, indicating a greater width than in other C. alpinus, similar to C. a. europaeus, and very close to larger individuals of C. priscus from Mosbach.

Comparing the dhole M1 talonid with those of other hypercarnivorous canids, it is possible to note different morphologies in the shape and position of the hypoconid. In Cuon, since the oldest specimens belonging to this genus, this cusp is conical-shaped and positioned approximately in the centre of the talonid basin. In the extant Lycaon and in the Pleistocene Xenocyon, this cusp is longitudinally flattened and crested, and positioned towards the labial side.

The second molar shows variable morphology. On the occlusal surface, the protoconid and, behind it, a small hypoconid, are present and aligned. The metaconid is absent in Equi IGF-7318V, poorly developed in San Sidero and well developed in Valserra. The San Sidero and Valserra specimens share the presence of two separated roots in the M2 (considered an archaic feature), while they are coalesced in the other second molars studied. In the

latter having the foramina more anteriorly located. In our opinion, the variability observed in fossil and living wolves suggests that this character should be treated with caution.

The mandibles from San Sidero and Buca del Tasso are larger than those from Valserra, and are close to the extant dhole in overall dimensions. As mentioned above, the lower fourth premolar has been considered very important in Cuon systematics. The small cusp described by Bourguignat (1875) in Cuon europaeus occurs in all of the P4s in the Italian specimens. The fourth lower premolar of Buca del Tasso is very large, close to Cuon priscus and at the top of the European dhole range, while the smaller P4 from Valserra lies at the bottom end of the range, close to Cuon alpinus europaeus and C. a. caucasicus (Fig. 3.a).

The lower carnassial teeth show typical Cuon features: the paraconid and protoconid are very well developed and sharp, the metaconid is reduced and, above all, there is a single conical-shape cusp, the hypoconid, in the centre of the talonid. The metaconid is reduced (clearly distinguished in the remains from Melpignano and Valserra), and crest-like, as seen in Buca del Tasso and Equi 7380V. In the M1 talonid, the posterior edge is well developed and two crests link the protoconid and the hypoconid, as can be observed in the Melpignano and Equi specimens. In contrast, these features are absent in the M1 from the Valserra and Buca del Tasso specimens. The first molars from the Buca del Tasso and Melpignano specimens are larger than those from Valserra and Equi.

Fig. 2 - Scattergrams of P4L/W (2.a) and M1L/W (2.b) on upper teeth of Italian and European dholes compared with other hypercarnivorous canids. See legend of fig. 3 for explanation of symbols. Measurements are in millimetres. Refer to Tab. 1 caption for explanation of teeth measurement abbreviations.

Fig. 3 - Scattergrams of P4L/W (3.a), M1L/M1TrL (3.b), M1L/W (3.c), M2L/W (3.d), M1L/M2L (3.e) on lower teeth of Italian and European dholes compared with other hypercarnivorous canids. See legend for explanation of symbols. Xenocyon antonii = Xenocyon falconeri; Cuon rosi, C. stehlini and Xenocyon africanus = Xenocyon lycaonoides (Rook, 1994; see also Martínez-Navarro & Rook, 2004; Moullé et al., 2006 and Sotnikova & Rook, 2010 for taxonomical debate). Measurements are in millimetres. Refer to Tab. 2 captions for explanation of teeth measurement abbreviations.

145M. Petrucci et alii - Italian Middle-Late Pleistocene Cuon

Bollettino della Società Paleontologica Italiana, 51 (2), 2012146

Aurelian faunal turnover (Sardella et al., 1998), although the presence of this canid cannot be confirmed for the older sites of Fontana Ranuccio and Colle Avarone.

The Italian Pleistocene dholes are similar in size and overall morphology to the Pleistocene European Cuon alpinus, with the exception of the fossil from Valserra, which is smaller and comparable in size to the extant species.

At present, the earliest Holocene record of Cuon at Riparo Fredian (layer 5) can be considered as the last occurrence datum of this carnivoran in Italy.

ACKNOWLEDGMENTS

We wish to express our gratitude to the people who provided access to the data and collections: Elisabetta Cioppi (Florence), Enrico Squazzini (Terni), Medica Assunta Orlando (Maglie), Rossella Carlini (MCZR-Rome), Fabio Parenti and Luca Bellucci (IsIPU-Rome), Ralf-Dietrich Kahlke and Dennis Roessler (Weimar), Ursula Goelich and Doris Naegel (Wien), P.-E. Moullé (Menton), I. Toro, M. Patrocino Espigares and S. Ros-Montoya (Orce). We would also like to thank Cristina Cilli (University of Turin), Chiara Sorbini (University of Pisa) and Eugenia Segre Naldini (IIPU-Rome) for the very useful information on the fossil remains they have collected and/or studied.

This work was supported by Research funds of Ateneo Federato Project, “La Sapienza” University of Rome: “Paleoecologia e biocronologia delle associazioni a mammiferi neogenico-quaternarie in Italia”, coordinator Raffaele Sardella. RS acknowledges 2008 - SYNTHESYS project AT 5028 (European Union-funded Integrated Activities grant): Early and Middle Pleistocene carnivores (Mammalia) of Europe - Naturhistorisches Museum Wien, Austria. MP acknowledges Post-Graduate Scholarship 2004 offered by “La Sapienza” University of Rome (stage at Malaga University), and PhD research grants of Dipartimento di Scienze della Terra, University of Turin (resp. G. Pavia).

Moreover, we gratefully acknowledge the insightful comments and helpful criticism of the manuscript by Brian R. MacKenzie, Jean-Philippe Brugal, Lorenzo Rook, Massimo Delfino e Danielle Schreve.

The work was communicated in the IX edition of SPI International Congress - Giornate di Paleontologia 2009, Apricena, winning the prize as best poster, offered by the Società Paleontologica Italiana.

REFERENCES

Abbazzi L., Arca M., Tuveri C. & Rook L. (2005). The endemic canid Cynotherium (Mammalia, Carnivora) from the Pleistocene deposits of Monte Tuttavista (Nuoro, Eastern Sardinia). Rivista Italiana di Paleontologia e Stratigrafia, 111 (3): 497-511.

Adam K.D. (1959). Mittelpleistozäne Caniden aus dem Heppenloch bei Guttemberg (Württemberg). Stuttgarter Beiträge zur Naturkunde, 27: 1-46.

Altuna J. (1983). Hallazgo de un Cuon (Cuon alpinus Pallas) en Obarreta, Gorbea (Vizcaya). Kobie, 13: 141-157.

Baryshnikov G. (1996). The dhole, Cuon alpinus (Carnivora, Canidae), from the Upper Pleistocene of the Caucasus. Acta zoologica cracoviense, 39 (1): 67-73.

Biddittu I., Cassoli P.F., Radicati Di Brozolo F., Segre A.G., Segre Naldini E. & Villa I. (1979). Anagni, a K–Ar dated Lower and Middle Pleistocene Site, Central Italy: preliminary report. Quaternaria, 21: 53-71.

Biddittu I. & Segre A.G. (1984). Colle Marino, Arce e Fontana Liri, Castro dei Volsci, Anagni-Fontana Ranuccio, Pofi-Cava Pompi, Ceprano, S. Giovanni Incarico-Lademagne, Pontecorvo. In AA VV. I Primi Abitanti d’Europa. De Luca ed.: 165-167.

Bologna P., Di Stefano G., Manzi G., Petronio C., Sardella R. & Squazzini E. (1994). Late Pleistocene mammals from

M2L/B scattergram (Fig 3.d), second lower molars from Valserra and San Sidero fall within the sample of Cuon cf. C. priscus. The M2 from Equi is smaller and close to Cuon alpinus fossilis, in the lowermost part of the graph.

The bifurcated roots seen in lower second molars seem to occur different times in the Late Pleistocene dhole, as testified to by the frequency of this feature in mandible samples described by Pérez Ripoll et al. (2010) from the Late Pleistocene site of Cova Negra (Valencia, Spain). In fact, even since the earliest studies (e.g., Mivart, 1890), variability of the lower second molar morphology has also been witnessed in the extant dhole.

The ratios M1L/M1TrL and M1L/M2L (resp. Figs 3.b and 3.e) are related to the degree of hypercarnivory (Bonifay, 1971). The first reveals that Valserra and Equi specimens are more similar to Lycaon pictus and Cynotherium sardous than Cuon alpinus, while the dholes from Melpignano and, above all, Buca del Tasso show a greater hypercarnivory, close to Cuon alpinus from Matuzka cave (Northern Caucasus, Russia) and to Xenocyon falconeri and Xenocyon africanus (=Lycaon lycaonoides Rook, 1994; Martínez-Navarro & Rook, 2003; see also Moullé et al., 2006 and Sotnikova & Rook, 2010). In particular, the ratio M1L/M2L reveals that the second molar of the individual from Valserra is well developed and lies closer to Cynotherium sardous than to any other European fossil dholes, while the M1L is comparable with the C. alpinus europaeus sample.

CONCLUSIONS

The occurrence of Cuon alpinus has been documented in the late Middle-Late Pleistocene of Italy, but its fossil record is very restricted compared to other canids, such as the Mosbach wolf or the grey wolf. On the basis of the faunal assemblages studied, niche overlap for the dhole and other canids was apparently high, since they were competing for the same medium-size ungulate prey, such as cervids, wild boar, and wild goats and sheep.

The morphological and morphometrical analyses on the Pleistocene dholes from Italy highlight a certain degree of variability, although a number of diagnostic features have been identified, supporting an attribution for all the specimens studied to the species Cuon alpinus.

Moreover, according to Pérez Ripoll et al. (2010) and Brugal & Boudadi Maligne (2011) the features of lower fourth premolar and first molar are consistent with attribution to the subspecies Cuon alpinus europaeus.

Taking into account the differences observed in the talonid basin and, in particular, in the hypoconid, it is possible to state that Cuon clearly differs from both Lycaon and Xenocyon, with the latter genus considered as its possible ancestor (Moullé et al., 2006; Tedford et al., 2009).

The second molar is variable both in shape and occlusal surface features in all the Italian dholes. The M2 variability observed in the extant dhole has been related to the latitudinal distribution of this canid (Mivart, 1890; Durbin et al., 2004). However, the paucity of the European Pleistocene dhole sample cannot provide detailed information on the condition for the fossil Cuon.

Cuon alpinus europaeus spread into Italy in the late Middle Pleistocene, possibly as part of the Galerian-

147M. Petrucci et alii - Italian Middle-Late Pleistocene Cuon

Girman D.J., Kat P.W., Mills M.G.L., Ginsberg J.R., Borner M., Wilson V., Fanshawe J.H., Fitzgibbon C., Lau L.M. & Wayne R.K. (1993). Molecular genetic and morphological analyses of the African wild dog (Lycaon pictus). Journal of Heredity, 84: 450-459.

Holliday J.A. & Steppan S.J. (2004). Evolution of hypercarnivory: the effect of specialization on morphological and taxonomic diversity. Paleobiology, 30 (1): 108-128.

Iyengar A., Babu V.N., Hedges S., Venkataraman B., Maclean N. & Morin P.A. (2005). Phylogeography, genetic structure, and diversity in the dhole (Cuon alpinus). Molecular Ecology, 14: 2281-2297.

Kurten B. (1968). Pleistocene mammals of Europe. 317 pp. Weidenfeld & Nicholson, London.

Lyras G.A., Van der Geer A.E., Dermitzakis M. & De Vos J. (2006). Cynotherium sardous, an insular canid (Mammalia: Carnivora) from the Pleistocene of Sardinia (Italy), and its origin. Journal of Vertebrate Paleontology, 26 (3): 735-745.

Malatesta A. (1962). Il cane selvaggio del Pleistocene di Sardegna. Geologica Romana, 1: 173-189.

Manzi G. (2004). Italian prehistoric promenades: the human fossil sample from the “earliest European” to the latest Neandertals. Zona Arqueológica, 4 (3): 221-231.

Martinez-Navarro B. (1992). Revisión sistemàtica y estudio cuantitativo de la fauna de Macromamiferos del yacimiento de Venta Micenea (Orce, Granada). In Gibert J. (ed.), Proyecto Orce-Cueva Victoria (1988-1992). Presencia humana en el Pleistoceno inferior de Granada y Murcia. Ayuntamiento de Orce, Granada (Spain): 21-85.

Martinez-Navarro B. & Rook L. (2003). Gradual evolution in the African hunting dog lineage. Sistematic implication. Comptes Rendus Palevol, 2: 695-702.

Mivart S.G. (1890). Dogs, jackal, wolves, and foxes: a monograph of the Canidae. 216 pp. R.H. Porter, London.

Moigne A.M., Palombo M.R., Belda V., Heriech-Briki D., Kamici S., Lacombat F., Lumley M.A., Moutoussamy J., Rivals F., Quiles J. & Testu A. (2006). Les faunes de grands mammifères de la Caune de l’Arago (Tautavel) dans le cadre biochronologique den faunes du Pléistocène moyen italien. Anthropologie, 110: 788-831.

Mottl M. (1951). Die Repolust-Höhle bei Peggau (Steiermark) und ihre eiszeitlichen Bewohner. Archaeologica Austriaca, 8: 1-78.

Moullé P.-E., Echaussoux A. & Lacombat F. (2006). Taxonomie du grand canidé de la grotte du Vallonnet (Roquebrune-Cap-Martin, Alpes-Maritimes). Anthropologie, 110: 832-836.

Nehring A. (1890). Uber Cuon alpinus fossilis nebst Bemerkungen über einige andere fossile Caniden. Neues Jahrbuch für Mineralogie, Geologie und Palaeontologie, 2: 34-52.

Nehring A. (1891). Diluviale Reste von Cuon, Ovis, Saiga, Ibex und Rupicapra aus Mähren. Neues Jahrbuch für Mineralogie, Geologie und Palaeontologie, 2: 107-155.

Pallas P.S. (1811). Zoographia Rosso-Asiatica. Tomus Primus. 568 pp. Petropoli.

Pérez Ripoll M., Morales Pérez J.V., Sanchis Serra A., Aura Tortosa J.E. & Sarrión Montañana I. (2010). Presence of the genus Cuon in Upper Pleistocene and initial Holocene sites of the Iberian Peninsula: new remains identified in archaeological contexts of the Mediterranean region. Journal of Archaeological Science, 37 (3): 437-450.

Pons Moyà J.,(1987). Los carnivoros (Mammalia) de Venta Micena (Granada, España). Paleontologia y Evoució. Memoria Especial, 1: 109-128.

Pons Moyà J. & Moyà Solà S. (1978). La fauna de Carnivoros del Pleistoceno medio (Mindel) de la cueva Victoria (Cartagena, España). Acta Geologica Hispanica, 13 (2): 54-58.

Rook L. (1994). The Plio-Pleistocene Old World Canis (Xenocyon) ex gr. falconeri. Bollettino della Società Paleontologica Italiana, 33 (1): 71-82.

Sardella R., Abbazzi L., Argenti P., Azzaroli A., Caloi L., Capasso Barbato L., Di Stefano G., Ficcarelli G., Gliozzi E., Kotsakis T.,

the Melpignano (LE) “Ventarole”: preliminary analysis and correlations. Bollettino della Società Paleontologica Italiana, 33 (2): 265-274.

Bonifay M.F. (1971). Carnivores quaternaires du Sud-Est de la France. Mémoires du Museum national d’Histoire naturelle, Serie C, 21: 145-153.

Boule M. & De Villeneuve L. (1927). La grotte de l’Observatoire a Monaco. Archives de l’Institut de Paléontologie Humaine, mémoire, 1: 1-112.

Bourguignat M.J.B. (1875). Recherches sur le ossements de Canidae constates en France à l’état fossile pedant la période quaternaire. Annales des Sciences géologiques, 6(6): 1-60.

Brugal J.-P. & Boudadi-Maligne M. (2011). Quaternary small to large canids in Europe: Taxonomic status and biochronological contribution. Quaternary International, 243(1): 171-182.

Cilli C., Malerba G. & Tozzi C. (1998). Analyse archeozoologique et modifications de surface des restes faunique de deux site du Paleolithique superior de Toscane (Italie). Atti U.I.S.P.P., Forlì, XIII, 2: 675-679.

Cipullo A., Pavia M., Petrucci M. & Sardella R. (2006). Nuovi resti di carnivori dalle “Ventarole” di San Sidero (Maglie, Lecce) (Pleistocene superiore). Abstract Volume Giornate di Paleontologia 2006, 8-11 Giugno 2006, Trieste: 20.

Colbert E.H. & Hooijer D.A. (1953). Pleistocene mammals from the limestone fissures of Szechwan, China. Bulletin American Museum of Natural History, 102: 1-134.

Cordy J.M. (1983). Découverte de Cuon alpinus europaeus Bourguignat dans le Quaternarie de Belgique. Mémoire de la Société Préhistorique de France, 16: 49-59.

Cotrozzi S., Mallegni F. & Radmilli A.M. (1985). Fémur d’un enfant néandertalien dans la Buca del Tasso a Metato, Alpi Apuane (Italie). L’Anthropologie, 89: 111-116.

De Giuli C. (1983). Le faune pleistoceniche del Salento. 1. La fauna di San Sidero 3. Quaderni del Museo di Paleontologia di Maglie, 1: 45-84.

De Grossouvre A. & Stehlin H.G. (1912). Les sables de Rosières, près Saint-Florent (Cher). Bulletin de la Société géologique de France, sér. 4 XII (3-4): 194-212.

Del Campana D. (1923) Sopra un Cuon e una Mustela del Quaternario di Equi (Alpi Apuane). Rendiconti della Reale Accademia dei Lincei Classe di Scienze fisiche, matematiche e naturali, vol. XXXII, ser. 5: 170-172.

Del Campana D. (1954). Carnivori quaternari della Tecchia e Caverna di Equi nelle Alpi Apuane. Palaeontographia Italica, 44: 4-21.

Driesch von den A. (1976). A guide to the measurement of animal bones from archaeological sites. 137 pp. Peabody Museum Bulletin, 1. Harvard University, Cambridge, MA.

Dubois A. & Stehlin H.G. (1933). La grotte de Cotencher, station moustérienne. Mémoires de la Société paléontologique suisse, 52-53: 1-292.

Durbin L.S., Venkataraman A., Hedges S. & Duckworth W. (2004). South Asia-South of the Hymalaya (Oriental). In Sillero-Zubiri C., Hoffman M. & Macdonald D.W. (eds), Canids: Foxes, Wolves, Jackals and Dogs, Oxford University Press, UK: 210-219.

Echassoux A., Moigne A.M., Moullé P.-E., Tianyuan L., Xiao Bo F., Wensen L. & Zianzhi W. (2008). Les faunes de grands mammifères du site de l’Homme de Yunxian (Quyuanhekou, Quingqu, Yunxian, Province du Hubei, République populaire de Chine). In de Lumley H. & Tianyuan L. (eds), Le site de l’homme de Yunxian (Quyuanhekou, Quingqu,Yunxian, Province du Hubei). CNRS Éditions, Paris: 253-364.

Fabiani R. (1923). La fauna mammologica quaternaria della Buca del Tasso. Atti del Comitato per le ricerche di Paleontologia umana in Italia. Archivio per l’Antropologia e l’Etnologia, 52 (1,4): 10-21.

García N. & Arsuaga J.L. (1997). The carnivore remains from the hominid-bearing Trinchera-Galeria, Sierra de Atapuerca, middle Pleistocene site (Spain). Geobios, 31 (5): 659-674.

Bollettino della Società Paleontologica Italiana, 51 (2), 2012148

Teppner W. (1914). Beiträge zur fossilen Fauna der steirischen Höhlen. Mitteilungen für Höhlenkunde, 7(1): 1-18.

Thenius E. (1954). On the origins of the dhole. Österreichische Zoologische Zeitschrift, 5: 377-388.

Tobien H. (1957). Cuon Hodgson und Gulo Frisch (Carnivora, Mammalia) aus den Altpleistozänen Sanden von Mosbach bei Wiesbaden. Acta Zoologica Cracoviensia, 2 (18): 433-451.

Trinkaus E., Marks A.E., Brugal J.P., Bailey S.E., Rink W.J. & Richter D. (2003). Later Middle Pleistocene human remains from the Almonda Karstic system, Torres Novas, Portugal. Journal of Human Evolution, 45: 219-226.

Van Valkenburg B. (1991). Iterative evolution of hypercarnivory in canids (Mammalia: Canidae): evolutionary interaction among sympatric predators. Paleobiology, 17: 340-362.

Wozencraft W.C. (1989). Classification of the recent Carnivora: 569-593. In Gittleman J.L. (ed.), Carnivore behavior, ecology and evolution. Cornell University Press, Ithaca, NY, USA.

Manuscript received 16 February 2012Revised manuscript accepted 25 September 2012Published online 28 september 2012Editor Edoardo Martinetto

Masini F., Mazza P., Mezzabotta C., Palombo M.R., Petronio C., Rook L., Sala B. & Torre D. (1998). Mammal faunal turnover in Italy from the Middle Pliocene to the Holocene. Netherlands journal of geosciences, 60: 499-511.

Schütt G. (1973). Revision der Cuon und Xenocyon (Canidae, Mammalia) aus den Altpleistozänen Mosbacher sanden (Wiesbaden, Hessen). Mainzer Naturwissenschaften Archiv, 12: 49-77.

Segre A.G. & Ascenzi A. (1984). Fontana Ranuccio: Italy’s earliest Middle Pleistocene hominid site. Current Anthropology, 25: 230-233.

Simpson G.G. (1945). The principle of classification and a classification of mammals. Bulletin of the American Museum of Natural History, 85: 1-350.

Sotnikova M. (2001). Remains of Canidae from the Pleistocene deposits of Untermassfeld. Römisch-Germanischen Zentralmuseums Mainz, 40: 607-632.

Sotnikova M. & Rook L. (2010). Dispersal of the Canini (Mammalia, Canidae: Caninae) across Eurasia during the Late Miocene to Early Pleistocene. Quaternary International, 212: 86-97.

Tedford R., Xiaoming W. & Taylor B. (2009). Phylogenetic systematics of the North American fossil caninae (Carnivora: Canidae). Bulletin of the American Museum of Natural History, 325:1-218.

Teilhard De Chardin P. (1940). The fossils from Locality 18 New Peking. Palaeontologica Sinica, C, 9: 11-14.