PALEVO 1631-0683 Elsevier S1631-0683(17)30107-0 10.1016/j.crpv.2017.08.004 Research article General Palaeontology, Systematics and Evolution (Vertebrate Palaeontology) General palaeontology, systematics and evolution Vertebrate Palaeontology The first French tragulid skull (Mammalia, Ruminantia, Tragulidae) and associated tragulid remains from the Middle Miocene of Contres (Loir-et-Cher, France) Le premier crâne de Tragulidae français (Mammalia, Ruminantia) et les restes associés de Tragulidae du Miocène moyen de Contres (Loir-et-Cher, France) Mennecart Bastien mennecartbastien@gmail.com a b de Perthuis Adrien c Rössner Gertrud E. d e Guzmán Jonathan A. d e f de Perthuis Aude c Costeur Loïc a Nathurhistorisches Museum Basel, Augustinergasse 2, 4001 Basel, Switzerland Nathurhistorisches Museum Basel Augustinergasse 2 Basel 4001 Switzerland Nathurhistorisches Museum Wien, Burgring 7, 1010 Vienna, Austria Nathurhistorisches Museum Wien Burgring 7 Vienna 1010 Austria 27, rue des Papegaults, 41000 Blois, France 27, rue des Papegaults Blois 41000 France SNSB–Bayerische Staatssammlung für Paläontologie und Geologie, Richard-Wagner-Strasse 10, 80333 München, Germany SNSB–Bayerische Staatssammlung für Paläontologie und Geologie Richard-Wagner-Strasse 10 München 80333 Germany Department für Geo- und Umweltwissenschaften, Paläontologie und Geobiologie, and GeoBio-Center der Ludwig-Maximilians-Universität München, Richard-Wagner-Str. 10, 80333 München, Germany Department für Geo- und Umweltwissenschaften, Paläontologie und Geobiologie, and GeoBio-Center der Ludwig-Maximilians-Universität München Richard-Wagner-Str. 10 München 80333 Germany Departamento de Ciencias Básicas, Universidad de Concepción, Campus Los Ángeles, Chile Departamento de Ciencias Básicas, Universidad de Concepción Campus Los Ángeles Chile 17 3 S1631-0683(18)X0004-9 189 200 © 2017 Académie des sciences. Published by Elsevier B.V. All rights reserved. 2017 Académie des sciences Full (PDF)

The Faluns Auger quarry (Contres, France) is famous for its Langhian mammal fauna (MN5) and tragulid remains have been identified as Dorcatherium sp. New excavations provided a fragmented skull of a tragulid, dental remains, and a metapodium which we describe here. Based on morphological and morphometrical characters, these specimens are attributed to Dorcatherium crassum. CT-scans provide insight in the petrosal bone and bony labyrinth. We prove intra-population variability of the p4 and intra-specific variability of the bony labyrinth. Nevertheless, we can demonstrate that the bony labyrinth is useful for systematics. Thus, we confirm the presence of D. crassum in the Pontlevoy–Thenay and Savigné-sur-Lathan Faluns Basins as reported already prior to our study. Yet, we cannot confirm evidence for Dorcatherium naui. D. naui specimens reported from these basins may belong to another species. D. naui should be absent from the Faluns before MN7, such as in the rest of Europe.

La carrière à faluns dite « Auger » à Contres (France) est connue pour avoir fourni une faune de mammifères du Langhien (MN5), dont un Tragulidae, identifié comme Dorcatherium sp. Lors de nouvelles excavations, un crâne partiel de Tragulidae, ainsi que des restes dentaires et un métapode ont été découverts. Sur la base de caractères morphologiques et morphométriques, ces derniers sont attribués à Dorcatherium crassum. Le pétreux et le labyrinthe osseux sont décrits. Il est mis en évidence qu’il existe une variabilité intrapopulationnelle de la p4, ainsi qu’une légère variabilité intraspécifique du labyrinthe osseux. Cependant, nous confirmons ici que cette structure interne peut être utilisée à des fins taxonomiques. D. crassum est déjà bien connu dans les bassins de Pontlevoy–Thenay et de Savigné-sur-Lathan. Nous soupçonnons cependant que les spécimens attribués à D. naui de ces bassins appartiennent à une autre espèce. D. naui est absent dans le reste de l’Europe avant MN7.

 Comparative anatomy, Langhian, Inner ear, 3D, Heterochrony Anatomie comparée, Langhien, Oreille interne, 3D, Hétérochronie presented Handled by Lorenzo Rook Introduction

The ruminant family Tragulidae, members of which commonly called chevrotains or mouse deer, includes the only living ruminant species that do not belong to the clade of the derived pecoran ruminants (Hassanin et al., 2012 and Nowak, 1999). They possess only three fully formed stomach chambers, missing the omasum (Boas, 1890, Clauss and Rössner, 2014, Milne-Edwards, 1864, Nowak, 1999 and Rössner, 2007) and are usually described as primitive (e.g., Métais and Vislobokova, 2007 and Rössner, 2007). Their origin in the Paleogene and their early evolution remain uncertain due to a sporadic fossil record and extremely few species and genera known during this period (Mennecart & Vislobokova; Mennecart et al., 2011, Métais and Vislobokova, 2007 and Métais et al., 2001; and references therein). In contrast, their Neogene fossil record is extensive in the Old World (e.g., Clauss and Rössner, 2014 and Gentry et al., 1999, Rössner, 2007, Sánchez et al., 2015). Over 30 species have been described in the Miocene, mainly from Asia (Kostopoulos and Sen, 2016 and Rössner, 2007). In Europe, their Neogene fossil record is exclusively affiliated to the genus Dorcatherium with seven different species from the latest early Miocene to the latest Miocene (D. guntianum, D. crassum, D. vindebonense, D. peneckei, D. naui, D. puyhauberti, and D. jourdani; Aiglstorfer et al., 2014, Rössner, 2007 and Rössner and Heissig, 2013). However, several authors (e.g., Rössner and Heissig, 2013 and Sánchez et al., 2015) recognize that Dorcatherium is most probably a polyphyletic genus.

European skull material, sufficiently well-preserved to allow α-taxonomic assessment, is currently only known from Germany (D. naui skulls from Eppelsheim described Kaup, 1839, the undescribed Dorcatherium skull from Steinheim am Albuch figured in Gentry et al., 1999: 23.3 [there affiliated to D. naui; but affiliated to D. guntianum in Rössner & Heissig, 2013 online resource 2 and to D. crassum in Aiglstorfer et al., 2014], and the superficially described D. crassum skull from Thierhaupten, Seehuber, 2015). Although, French localities also yielded an abundant fossil record for Dorcatherium, from none of those a skull or partial skull has been reported up to now, not even from the outstanding Sansan fossil site (Gers), which is the type locality for D. crassum (Morales et al., 2012). Yet, since D. crassum is the only tragulid species known from Sansan (Mennecart and Costeur, 2016b and Morales et al., 2012) attributed an isolated petrosal bone from Sansan to it (see Mennecart & Costeur, 2016c for access to 3D reconstruction).

Here we provide the description of a partial skull and other dental and postcranial remains from the Auger quarry area in Contres (Loir-et-Cher, Pontlevoy–Thenay Basin) ca. 200 km south-southwest of Paris (Fig. 1) attributed to D. crassum. This is the first largely preserved tragulid skull known from France and the first extensive description for a skull of D. crassum.

 Material & methods Geological setting

Castillo et al. (2006) and Gagnaison et al. (2006) already mentioned some remains of Dorcatherium sp. from the Auger quarry. The site was exploited for fossils from 1995 to 1997 (Gagnaison et al., 2006). It is now totally exploited and the locality is sealed, but from 2012 to 2016, a last excavation led by two of us provided the new D. crassum material described here. The fossils have been discovered in an estuarine layer overlaying the fully marine Falun deposits (Castillo et al., 2006). The abundant fauna (more than 1.000 specimens discovered during two several-weeks lasting excavation campaigns) permits an assignation to the sub-mammal zone MN5b, ca. 15Ma (Castillo et al., 2006).

 Material

The original specimens are currently stored in the private collection of A. de Perthuis. These specimens will be given to a natural history museum and all the data about the location of these fossils (currently in the collection or to be collected in the future) will be provided to the Naturhistorisches Museum Basel, along with casts to facilitate management of the collection. Casts and digital specimens are kept in the Naturhistorisches Museum Basel (Switzerland). Digital 3D data of the skull (AP2014.0715 and cast NMB Fa.213.abg), its isolated right petrosal, and the reconstructed left bony labyrinth are freely accessible on the MorphoMuseum web site (Mennecart et al., online). Specimens of recent tragulids for comparison (Hyemoschus, Moschiola, and Tragulus) are stored in the Naturhistorisches Museum Basel.

To access the bony labyrinth and create a virtual model of the skull (3D data made of the maxilla and associated teeth, the basicranium, excepted for the isolated part of basisphenoid and basioccipital, and the skull roof only), the specimens were scanned using high-resolution X-ray computed tomography (Phoenix Nanotom, GE, at the Biomaterials Science Centre of the University of Basel, Switzerland). Scanning resolution was 55 μm for the entire skull and 35 μm for a second scan focusing on the ear region. A total of 1440 equiangular radiographs were taken over 360° using an accelerating voltage of 170 kV and a beam current of 30 mA. Segmentation of the specimen was carried out with the AVIZO 9.0 software (Visualization Sciences Group).

 Terminology

Terminology for skull elements follows Baron (1999), for the petrosal bone O’Leary (2010), for the bony labyrinth Ekdale (2013), for tooth crown elements Bärmann & Rössner (2011) (see Fig. 2) and for postcranial elements Köhler (1993). Measurements have been taken with a calliper (0.1 mm precision).

 Abbreviations

NMB: Naturhistorisches Museum Basel; AP: collection A. de Perthuis; p: lower premolar; m: lower molar; P: upper premolar; M: upper molar; Mt: metatarsal bone.

 Systematic Paleontology and description

Order: Artiodactyla Owen, 1848

Suborder: Ruminantia Scopoli, 1777

Infraorder: Tragulina Flower, 1883

Family: Tragulidae Milne Edwards, 1864

Genus Dorcatherium Kaup, 1833

Dorcatherium crassum (Lartet, 1839)

Fig. 2, Fig. 3, Fig. 4, Fig. 5 and Fig. 6 Table 1, Table 2 and Table 3.

Referred specimens. AP2014.0715 (NMB Fa.213.abg digital copy, see Mennecart et al., 2017, and 3D print) fragmented skull containing the basicranium represented by isolated basisphenoid and basioccipital, the skull roof, and the left maxilla with P4–M3 mainly (Fig. 3 and Fig. 4); AP2013.0135 right maxilla with P4–M3; AP2015.0134 right maxilla with M1–M3; AP2012.0130 (NMB Fa.214.abg cast) right corpus mandibulae with p3–m3; AP2014.0129 (NMB Fa.215.abg cast) right corpus mandibulae with p2–m3 (Fig. 5.1); AP2015.0131 left m2–m3; AP2015.0133 fragmented right corpus mandibulae with p3–m1; AP2013.0132 fragmented right corpus mandibulae with p3 and alveolus of two rooted p2; AP2015.0128 (NMB Fa.216.abg cast) left Mt III–IV (Fig. 5.2).

Skull. The skull is highly fragmented but some delicate structures (e.g., eye socket, bulla, and probably vomer) are partly conserved. The best-preserved bones are from the skull roof (frontals, parietals, squamosals, proximal part of the nasals, and the left maxilla bone), and the basicranium (basisphenoid, and the occipital). The naso-frontal suture is W-shaped, with the posterior part of it, reaching the level of the centre of the orbits in dorsal view. The frontals form an elongated bone extension going between the nasal bones. In living species, the nasal and frontals suture is also not straight, but is farther anterior of the orbits and relatively flat, although a little variation is observed within a species (pers. obs. on Tragulus kanchil and Hyemoschus aquaticus) with individuals fluctuating from flat to convex shape nasal bone. The supraorbital foramen is larger than in living the tragulids. It is separated from the deep supraorbital groove by a robust bony bridge. These structures are slender in living tragulids (Rössner, 2007: fig. 16.3A; pers. obs.; Smit-van Dort, 1989: fig. 1). The orbit is well-rounded such as in D. naui from Eppelsheim and D. crassum from Steinheim am Albuch. However, this character is highly variable in living tragulids, being rounded, rectangular, or even more triangular. The postorbital bar of the frontal process is broad on the orbital face, such as in the Thierhaupten specimen. In living tragulids, this part is more slender. The sagittal crest is well-developed separating into the left and the right parietofrontal suture in the middle of the braincase with an angle of 90°, such as in the Thierhaupten specimen. It is a wide ridge and passes at the dorsocaudal skull edge to the well-developed nuchal crest forming the framing external crests of the occipital. Among the living tragulids, only Hyemoschus possesses such structures, but more weakly developed. In Tragulus and Moschiola, the sagittal crests are incipient and the nuchal crest is weak to absent (Rössner, 2007: fig. 16.3A; Smit-van Dort, 1989: fig. 1). Contrary to Hyemoschus, no external occipital crest is observed in the Doractherium specimen described here. Similar to Hyemoschus, the dorsal part of the occipital condyles is somewhat detached from the dorsal condylar fossa area. The two nuchal tubercles are well-developed, while they are absent in living taxa. The jugular processes, even if broken, are more similar in shape to Hyemoschus than to the other living tragulid genera Moschiola and Tragulus in being more profiled and relatively larger. It is even stockier than in Hyemoschus and possesses a tuberosity.

Petrosal bone. The petrosal bone has a rather broad epitympanic wing with a pointed apex, it is more elongated than in Tragulus and Moschiola (Mennecart & Costeur, 2016b). The promontorium on the ventrolateral side is tubular in shape with a broad posterior part (Fig. 4.a1). It has a flat convex surface like the petrosal bone of D. crassum from Sansan (Mennecart & Costeur, 2016b) and does not show the two distinct convexities seen in Tragulus (O’Leary, 2010). There is no apparent transpromontorial sulcus as already observed on all living tragulids and on D. crassum from Sansan (Mennecart and Costeur, 2016b and O’Leary, 2010). The fossa for the tensor tympani muscle is large as in Hyemoschus and has a quadratic shape that becomes more elongated anteriorly. It does not excavate into the tegmen tympani. It occupies slightly more than half the size of the promontorium, much of its anterior part. The crista interfenestralis between the fenestrae vestibuli and cochleae is broad and both fenestrae have about the same size. A small but deep fossa for the attachment of the stapedial muscle is visible directly posteriorly to the fenestra vestibuli. The facial sulcus is nor very broad neither very deeply impressed. No protruding articulations to the medial or posterior bulla are visible, but this may be due to the 3D reconstruction. In contrast the articulation to the anterior bulla is well visible on the ventrolateral side next to the anterior process of the tegmen tympani.

The dorsomedial side shows a knob (Fig. 4.a1), again as in Hyemoschus and unlike in Tragulus or Moschiola. The hiatus fallopii sits within a subcentral elongate slit of the tegmen tympani. The tegmen tympani extends anteriorly almost up to the level of the most anterior part of the promontorium, being slightly smaller than on the specimen from Sansan. It has a blunt apex. No basicapsular groove is visible recalling the situation in Sansan. It is also hard to observe on the three living tragulid genera as shown by Mennecart & Costeur (2016b). The subarcuate fossa is rather shallow. The opening of the vestibular aqueduct sits centrally posterior to the internal acoustic meatus as on the specimen from Sansan. The opening for the cochlear aqueduct is located more anteriorly on the ventromedial side at the level of the internal acoustic meatus and it sits on the dorsal extension of the ventromedial side. The mastoid region is wedge-like as in living and fossil tragulids where it is known and as in ruminants in general (Mennecart and Costeur, 2016b and O’Leary, 2010).

Bony labyrinth. The bony labyrinth is very similar to that of D. crassum described in Mennecart & Costeur (2016b). Linear and volumetric measurements are given in Table 1. The cochlea has three turns (Fig. 4.b2). It is massive, compact and tall because of the large number of turns compared with other ruminants (Costeur, 2014). The aspect ratio (total height divided by total width) is 0.55, slightly below those of the living tragulids and of D. crassum from Sansan (Mennecart & Costeur, 2016b) but in the range of high cochleas. The cochlear spiral is tight and only half of the basal turn is detached from the rest of the cochlea. The cochlear aqueduct is enlarged, with a flattened diameter. The secondary bony lamina is well visible over most of the basal turn, so apparently longer than in D. crassum from Sansan, but preservation could play a role here since the specimen from Sansan was recrystallized (Mennecart & Costeur, 2016b) and thus less well preserved than here. The fenestra cochleae is rounded in shape and faces laterally. On the vestibule, both the sacculus and utriculus are well defined (Fig. 4.b1) such as on the Sansan specimen. The fenestra vestibuli is large and oval in shape, with a stapedial ratio of 1.62, exactly like D. crassum from Sansan. The anterior semicircular canal is the most dorsally expanded canal; it expands largely above the level of the common crus, as for the posterior semicircular canal but to a lesser extent for the latter. The common crus is less indented and longer than on the specimen from Sansan (Mennecart & Costeur, 2016b) but such a variation was observed on the living Tragulus kanchil (Mennecart & Costeur, 2016a). The lateral semicircular canal is straight. It branches dorsally above the posterior ampulla, in the vestibule between the latter and the base of the common crus, a typical feature of Tragulidae. In addition and in line with previous observations on extinct and extant tragulids, the lateral semicircular canal extends beyond the plane of the posterior semicircular canal in dorsal or lateral view. The vestibular aqueduct is a thin canal detached from the common crus over most of its course, as in the Sansan specimen. It originates in the vestibule at the base of the common crus but anteriorly. It runs parallel to the midline of the common crus along its course and is thus not curved; it extends above the end of the common crus dorsally, less so on this specimen than on that of Sansan because the common crus is longer here. The endolymphatic sac at the end of the vestibular aqueduct is large and broad, triangular in shape with a broad base, larger than on the Sansan specimen, but this feature is ontogenetically variable in size.

Upper dentition. The enamel is wrinkled. The cusps are bunoselenomorph. The triangular P4 possesses only one labial cone with an extremely little developed labial rib. The worn anterolabial crista joins the bulbous base of the anterior style that has a crescent shape and forms the anterolabial edge of the tooth. This style is well-developed. The anterolingual crista is straight and links the anterior style to the medial and crescentic lingual cone. The posterolingual crista is short, stopping at the level of the middle of the posterolabial crista. The posterolabial crista is a little concave and joins the posterolabial edge of the tooth. There is no marked posterior style. The lingual cingulum is well marked from the anterolingual edge of the tooth to its posterolabial edge. The upper molars increase in overall size from M1 to M3 (Table 2). They are relatively square in shape. The paracone rib, as well as mesostyle and parastyle are well-developed in contrast to the weak metacone rib. They increase in size from M1 to M3 as well. Paracone and metacone are almost in line. Labial cristae are short and straight. The metastyle is only present on M3. The anterior fossa is widely opened due to a postprotocrista posteriorly oriented. The straight preprotocrista is interrupted by the anterolabial edge of the anterior tooth. The metaconule is V-shaped. The premetaconulecrista is straight and ends in between the paracone and metacone. The postmetaconulecrista is incomplete on M1 and does not reach the posterolabial edge of the tooth. The posterior cingulum is weak on M1, is weaker on M2, and is absent on M3. The anterolingual cingulum becomes more and more pronounced from M1 to M3. There is no entostyle or accessory fold.

Mandible. The corpus mandibulae gradually enlarges in height from rostral to caudal below the teeth until the concavity of the incisura vasorum starts. This incisura is not marked, but smoothed. The diastema starts right in front of the p2, similarly to the symphysis. The small and oval mandibular foramen is at the level of the tooth row. The two mental foramina are on the diastema and form an elongated fissure since they are connected.

Lower dentition. The lower premolars are more or less of the same size (Table 2). The p2 and p3 have a similar shape. These teeth are composed of three main conids aligned in the axis of the tooth. The mesolabial conid is central and the highest conid in all the premolars. All the cristids are straight, the laterally compressed anterior conid being anteriorly oriented in p2 and p3. The posterolabial conid bears a weakly developed curved posterior stylid that forms the posterolabial edge and the posterior part of the p2 and p3. The p3 differs from p2 in having a minute posterolingual conid absent on the other tooth. The p4 bears a posteriorly oriented transverse cristid. The posterior stylid is big and partly closes the steep-sided posterior valley between the transverse cristid and the posterolabial cristid. The anterior part of the p4 may bear an anterior stylid (AP2014.0.129) or not (AP2013.0130). The molars are bunoselenodont with a bulky lingual wall. They increase in size from m1 to m3. The preprotocristid is highly curved and joins the premetacristid on the anterior part of the teeth. The other cristids are straight. The external postprotocristid joins the prehypocristid, while the internal postprotocristid fuses with the internal postmetacristid forming a “Tragulus fold”. The external postmetacristid forms a true “Dorcatherium fold”. The association of the “Tragulus fold” with the “Dorcatherium fold” forms the typical tragulid “M-structure”. The entoconid is a little oblique, giving birth to a straight preentocrisitd that fuses with the distal part of the internal postmetacristid and postprotocristid in the centre of the occlusal area. The lingual entoconid ridge on the apex of the enctonid is absent on m1, tiny on m2, and small on m3. The postentocristid is short and does not reach the base of the posthypoconidcristid. Since the entoconid is very bulky, there is an internal groove between this cristid and the cuspid. The relatively high hypoconulid forms the posterolabial corner of the m3. The prehypoconulicristid fuses with the base of the posthypocristid while the oblique back fossa of the m3 is opened lingually. The ectostylid is weakly developed and the posterior ectostylid is absent. The anterior cingulid is increasingly developed from m1 to m3. There is neither a metastylid nor an entostylid.

Postcranial bones. The metatarsal bones III and IV are fully fused in AP2015.0128 (size data are provided in Table 3). The proximo-plantar area is damaged. The distal part is enlarged compared to the diaphysis and keels articulating with the first phalanges are absent dorsally and distally. 22 mm above the distal end of the bones, an elongate, enlarged and deep metacarpal sulcus runs from dorsal to plantar. There are no additional crests on the edge of the sulcus (similar to the fig. 19 of Köhler, 1993). From the lateral view, the distal articular surfaces are flattened. The verticilli are weak and continue on the plantar surface in straight lines forming no real separation between the keels and the metaphysis. At the level of the greatest transverse diameter, there is a crest separating the rounded dorsal part of the bones from the flattened plantar one. The metatarsal bones are somewhat concave distally on the plantar side and convex on the dorsal side. On the proximal part of the bones, there is a longish furrow for the extensor tendon. On the plantar side, there are two small furrows which formed the articular areas with Mt II and Mt V. The proximal view does not allow a precise description of the articular surfaces. Nevertheless, it seems that the cuneiform articular surface of Mt III is half-moon shaped. The Mt III extends farther proximad, while the Mt IV extends farther distad. The metatarsal bones of D. crassum described herein are characteristic of Köhler's type A2 (wooded, very humid to semiaquatic habitat; Köhler, 1993).

 Comparison and discussion Taxonomical attribution

The p4, lacking a mesolingual conid, is typical for Tragulina (Mennecart, 2012 and Mennecart et al., 2011). Moreover, the “M-structure” formed by the presence of a “Tragulus” fold and a “Dorcatherium” fold and the general shape of the bony labyrinth including the insertion of the lateral canal in the vestibule high dorsally between the posterior ampulla and the common crus associated with the high number of cochlear turns are unique to members of Tragulidae (Costeur, 2014, Costeur et al., 2014, Gentry et al., 1999, Mennecart et al., 2011, Mennecart and Costeur, 2016a, Mennecart and Costeur, 2016b, Mennecart and Métais, 2015 and Métais and Vislobokova, 2007). All European tragulid species known are attributed to Dorcatherium (e.g., Aiglstorfer et al., 2014, Gentry et al., 1999 and Rössner, 2007, Rössner & Heissig, 2013). Van der Made (1996) mentioned a cf. Dorcabune anthracotherioides in the late Miocene of Kastellios K east (Crete, Greece). Dorcatherium clearly differs from Paleogene members of Tragulidae (Archaeotragulus and Iberomeryx) and the Neogene Afrotragulus notably in size (being much larger), the proportional length of the diastema, and the more brachydont condition (Mennecart, 2012, Mennecart et al., 2011, Métais et al., 2001 and Sánchez et al., 2010). But most probably Dorcatherium is a polyphyletic genus (Rössner and Heissig, 2013 and Sánchez et al., 2015) and thorough revision is needed (Rössner, 2007). Two main “lineages” or “ecomorphotypes” can be separated among European Dorcatherium: the bunoselenodont (D. crassum, D. vindebonense, D. peneckei) and the selenodont (D. guntianum, D. naui, D. puyhauberti, and D. jourdani) (Aiglstorfer et al., 2014, Fahlbusch, 1985, Morales et al., 2012, Mottl, 1961, Rössner, 2007 and Rössner and Heissig, 2013). Moreover, contrary to Mennecart et al. (2011), we want to point out here that not all species attributed to Dorcatherium do possess a p1. The presence or absence of a p1 may vary within a population (e.g., Aiglstorfer et al., 2014). Nevertheless, D. vindebonense and D. naui very often possess a p1 (Aiglstorfer et al., 2014, Alba et al., 2011 and Ginsburg and Mornand, 1986 for D. naui), while D. crassum mostly lacks this tooth, as it is part of its diagnosis (Aiglstorfer et al., 2014, Ginsburg and Mornand, 1986 and Morales et al., 2012; but see Rössner, 2010: fig. 1N and Gentry et al., 1999: fig. 23.2, there mistakenly named D. naui). The two mental foramina are located at the level of the diastema and form an elongated fissure since they are connected in D. crassum (Morales et al., 2012). They are well separated in D. naui (Aiglstorfer et al., 2014) being placed below p1 and p2 and below the middle of the diastema. Aiglstorfer et al. (2014) noticed that the more bunodont Dorcatherium species (D. crassum and D. vindebonense) possess simpler p4 structures than D. naui and D. guntianum, which have a complex structure in the posterior valley. Moreover, in the selendont Dorcatherium species (D. naui. and D. guntianum), the third lobe of the m3 is turned to labial, while in the bunodont ones (D. crassum, D. vindebonense, and D. peneckei), the third lobe is not turned as much (Aiglstorfer et al., 2014). The structure of the p4 and m3 of the studied specimens supports that they belong to the bunoselenodont “lineage”.

Additionally, Dorcatherium species may also be separated by size (Aiglstorfer et al., 2014, Alba et al., 2011, Fahlbusch, 1985, Morales et al., 2012 and Rössner and Heissig, 2013). Within Dorcatherium, D. vindebonense, and D. peneckei remains are large to very large, while those of D. guntianum and D. puyhauberti are small. Those of D. naui, D. jourdani, and D. crassum are medium-sized. The size dimensions of the studied specimens are similar to those of D. crassum from Sansan (Morales et al., 2012), Thierhaupten (Seehuber, 2015), and Sandelzhausen (Rössner, 2010). Moreover, since our specimens are diagnosed bunoselenodont medium-sized Dorcatherium specimens (see Table 2) that lack a p1, we have every reason to attribute them to D. crassum. Moreover, the metatarsal bones of D. crassum described by Rössner (2010) and Morales et al. (2012) are similar in shape and size (see Table 3).

 Intraspecific variability in <italic>D.</italic> <italic>crassum</italic>

We noticed some differences in the structure of the p4 in our set of specimens (Fig. 6). AP2014.0.129 possesses a p4 that is a little more derived with the presence of an anterior stylid, absent from the other specimens and from the specimens from Sansan (see figures in Morales et al., 2012). However, Aiglstorfer et al. (2014) and Mennecart (2015) already noticed that the p4 may be variable within populations in Tragulina (being more or less molarized, bearing additional cristids and folds) what corresponds to findings in other ruminants (Janis & Lister, 1985) and classical intraspecific variability. In our specimens, the differences may also be linked to difference in the preservation.

Rössner (2007) and Morales et al. (2012) proposed to use the fusion degree of the bones to better understand the evolution of Tragulidae. Hyemoschus possesses less fused hind limb bones than its Asian (and more derived; Mennecart and Costeur, 2016b and Sánchez et al., 2015) relatives (Rössner, 2007). Associated to dental characters, it has been proposed that Dorcatherium and Hyemoschus could be congeneric (see in Gentry et al., 1999). However, Morales et al. (2012) noticed that the grade of fusion of the metatarsal bones varies a lot within the Sansan population. It ranges from unfused to well-fused (without noticeable bone suture) in adult specimens. So, this character should be taken with caution. The here described specimen AP2015.0128 is fully fused without any noticeable bone suture.

It is commonly admitted that the morphology of the ear region can be correlated with ecological and environmental parameters (e.g., Billet et al., 2012, Grohé et al., 2015 and Pfaff et al., 2015). However, the bony labyrinth and ossicles are proven to fully ossify in foetuses long before birth at mid-gestation time in some ruminants (Costeur et al., 2016a, Costeur et al., 2016b and Mennecart and Costeur, 2016a). In addition, the morphological intraspecific variability of the bony labyrinth, observed within Artiodactyla, is smaller than the interspecific morphological diversity and is thus highly conservative (Aguirre-Fernández et al., 2017 and Mennecart and Costeur, 2016a). Thus, recent studies propose that the ear region supports a strong phylogenetic signal (Ekdale, 2013, Mennecart and Costeur, 2016b and Mennecart et al., 2016). The bony labyrinth of D. crassum described by Mennecart and Costeur, 2016b and Mennecart and Costeur, 2016c is very similar to the specimen described here even if the shape and length of their common crus are different. Mennecart & Costeur (2016a) already noticed intraspecific variation of this structure in living Tragulidae. The height of the common crus may form a more or a less acute angle between the anterior and the posterior canal and accordingly influences the relative length of the vestibular aqueduct to the common crus height (Mennecart & Costeur, 2016a). Here we confirm the specific attribution of the isolated petrosal bone described by Mennecart & Costeur (2016b) to D. crassum and we demonstrate that the bony labyrinth can be used confidently for taxonomic purposes.

 Skull heterochrony in Tragulidae

Developmental heterochrony is known in ruminant lineages. The development of the appendages in Cervidae due to social competition (hypermorphs, Geist, 1998) may be related to peramorphosis. Vrba et al. (1994) highlighted for an antelope lineage (Reduncini) that paedomorphosis (may be neoteny) affected cranial and reproductive characters and occurred together with an increase of the body size within the genus Kobus. Van der Geer (2005) suggested that insular artiodactyls that are exhibiting ancestral characters could be considered as paedomorphic dwarfs.

We noticed that D. crassum skull characters are similar to those of Hyemoschus (e.g., occipital condyles somewhat detached). However, the characters often are hyper-developed in D. crassum, while they are much less developed in Tragulus and Moschiola, Hyemoschus being intermediary (sagittal crest, nuchal crest, jugular process; see Table 4). It is interesting to note that these crests are also weaker in Hyemoschus juveniles and female. Life history characters in Hyemoschus, on the one hand, and in Moschiola plus Tragulus, on the other hand, are totally different. Hyemoschus weights two to eight times the weight of Moschiola and Tragulus. Moreover, the gestation time (ca. 150 days) and the sexual maturity (ca. 5 months) in Moschiola and Tragulus is very short in comparison to Hyemoschus (225 days and 18 months respectively; see Dubost et al., 2011, Table 4). Though, all three genera have a similar age at weaning (ca. 10 to 12 weeks) and a similar potential longevity (12–15 years).

According to phylogenetic analyses (Mennecart and Costeur, 2016a, Mennecart and Costeur, 2016b, Mennecart and Costeur, 2016c and Sánchez et al., 2015), Hyemoschus is the sister group of the clade composed by Tragulus and Moschiola, while Dorcatherium is the sister taxon of the three living genera. Hence, above described differences between these taxa may indicate a paedomorphosis process, probably a progenesis, in development and in reproduction strategies during tragulid evolution. Body size reduction associated with progenesis may “have been selected for by high environmental unpredictability and might also be correlated with dietary changes in the small species” (Bärmann, 2014) that, in the case of tragulids, is based practically only on fruit (Heydon & Bolloh, 1997). Another explanation for the body size reduction is the diversification of the more efficient in fiber consumption Pecora groups that displaced to tragulids to niches in which a frugivorous diet is possible year-round (Clauss & Rössner, 2014).

 Presence of <italic>Dorcatherium</italic> in the Faluns Formation

Dorcatherium is well-known to occur abundantly in the sediments of the Pontlevoy–Thenay Basin (Loir-et-Cher, France) from where D. crassum specimens have often been reported (e.g., Collier and Huin, 1977, Ginsburg and Mornand, 1986, Mayet, 1908 and Stehlin, 1925). The Faluns Formation (Blaisois, Pontlevoy–Thenay Basin; Touraine, Savigné-sur-Lathan Basin) is mainly composed of marine sand deposits (Ginsburg, 2001 and Ginsburg and Mornand, 1986) and famous for its paleontological content. No less than 58 species of mammals indicating European Neogene Mammal Unit MN5 are currently known from the Faluns (Ginsburg, 2001 and Ginsburg and Mornand, 1986), which are clearly distinct from older and reworked specimens (eroded, broken, darker colour), characteristic for MN2 (“calcaire de Beauce” Formation), MN3, and MN4 (“sable de l’Orléanais” formation), or even extremely rare dinosaur remains, which can be found there as well (Buffetaut and Pouit, 1994 and Ginsburg, 2001). Of the MN5 mammal community, Dorcatherium is particularly well-documented in the Faluns deposits. In 1925, Stehlin already reported on specimens of D. crassum, D. guntianum, and D. rogeri (= D. vindobonense) in the Bourgeois collection (Pontlevoy, Pontlevoy–Thenay Basin) and Ginsburg (2001) on D. crassum, D. guntianum, and D. peneckei in the Basin of Savigné-sur-Lathan. However, until now, no skull was known.

These data match the evidence from other European sites that D. crassum and D. guntianum, appear in Europe within MN4 during late Burdigalian (late early Miocene), and that D. peneckei, and D. vindobonense appear in Europe with MN5 during Langhian (early middle Miocene) (Aiglstorfer et al., 2014, Rössner, 2017 and Rössner and Heissig, 2013). Oddly, the presence of D. naui in the Faluns from Touraine and Blaisois (Langhian) was proposed based on a unique isolated tooth (Gagnaison et al., 2012) or in faunal lists (Gagnaison, 2013 and Gagnaison et al., 2009). This is in conflict with data from the rest of Europe, that document that D. naui appeared in Europe not earlier than the Serravalian, within MN7 (Aiglstorfer et al., 2014), i.e. 3 million years later than D. crassum. Ginsburg (2001) also published faunal lists from the same localities. It is interesting to note that there is no referring to D. naui, but instead D. guntianum is listed (Ginsburg, 2001). Moreover, the isolated m3 attributed to D. naui from the falunière de Tourelet has the “morphological characteristics of Dorcatherium guntianum” (Gagnaison et al., 2012), being crescentic. Both species belong to the selenodont Dorcatherium morphotype. However the size of the tooth is too small to belong to D. naui (12.9 × 7.3 mm) and falls in the range of D. guntianum observed by Aiglstorfer et al. (2014). Thus, we suppose that the D. naui specimens reported on by Gagnaison in his several studies correspond to D. guntianum. Yet, Ginsburg & Mornand (1986) described and illustrated specimens of D. cf. naui in younger Faluns deposits. Indeed, a second transgressive event is registered in the Doué-la-Fontaine Basin dated to MN9 (Vallesian) where these fossils were found.

It is noteworthy that the other Dorcatherium specimens from the falunière de Tourelet (Gagnaison et al., 2012), that are attributed to D. guntianum, are too big to belong to this species and the proportions do not correspond to tragulid lower molars (m1 28.0 × 24.0 mm and m2 22.0 × 25.0 mm).

 Conclusion

This is the first description of a fossil tragulid skull found in a French locality. It provides access to internal skull structures which are associated with reliably systematically assignable dental remains in a fossil tragulid for the first time. The description of the internal morphology of the ear region corresponds to Mennecart & Costeur (2016b) description of an isolated petrosal bone from the Sansan site to D. crassum. Hence, our work supports that the bony labyrinth solely can be used confidently for systematic classification at species level. The well-preserved material also sheds light on the intraspecific variability (teeth and bony labyrinth) within a fossil tragulid species and population. The p4 complexity, as well as the common crus of the bony labyrinth, may vary.

 Acknowledgements

All our gratitude goes to Mr. Perceval, the owner of the field, who kindly gave all the authorizations for digging. Furthermore, we thank the three reviewers (Maeva Orliac, Manuela Aiglstorfer, and Reviewer3) as well as the editor Lorenzo Rook for their very helpful comments, which largely improved the manuscript. The Swiss National Foundation is warmly thanked for granting the SNF Projects 200021_159854/1, P300P2_161065, and P3P3P2_161066.

 Aguirre-Fernández G. Mennecart B. Sánchez-Villagra M.R. Sánchez R. Costeur L. A fossil dolphin ear bone from the northern Neotropics, insights into habitat transitions in iniid evolution J. Vert. Paleontol. 2017 10.1080/02724634.2017.1315817 e1315817 Aiglstorfer M. Rössner G.E. Böhme M. Dorcatherium naui and pecoran ruminants from the late Middle Miocene Gratkorn locality (Austria) Palaeobiodivers. Palaeoenviron. 94 2014 83–123 Alba D.A. Moyà-Solà S. Robles J.M. Casanovas-Vilar I. Rotgers C. Carmona R. Galindo J. Middle Miocene tragulid remains from Abocador de Can Mata: the earliest record of Dorcatherium naui from Western Europe Geobios 44 2011 135–150 Bärmann E.V. The evolution of body size, horn shape and social behavior in crown Antilopini – an ancestral character state analysis Zitteliana B 32 2014 185–196 Bärmann E.V. Rössner G.E. Dental nomenclature in Ruminantia: Towards a standard terminological framework Mamm. Biol. 76 6 2011 762–768 Baron R. Anatomie comparée des mammifères domestiques : Tome 1. Ostéologie 4e ed. 1999 Vigot Paris 761 p Billet G. Hautier L. Asher R.J. Schwarz C. Crumpton N. Martin T. Ruf I. High morphological variation of vestibular system accompanies slow and infrequent locomotion in three-toed sloths Proc. R. Soc. B 279 2012 3932–3939 Boas J.E. Zur Morphologie des Magens der Cameliden und Traguliden und über die systematische Stellung letzterer Abtheilung Morphol. Jahrb. 16 1890 494–524 Buffetaut E. Pouit D. Dinosaur and crocodilian remains from the Upper Cretaceous of west-central France C. R. Acad. Sci. Paris Ser. II 319 2 1994 253–259 Castillo L. Gagnaison C. Grugier O. Renou J.-C. Étude paléoécologique du site miocène de Contres (L.-et-Cher, France) Bull. Soc. Etudes Sci. d’Anjou XX 2006 61–78 Clauss M. Rössner G.E. Old world ruminant morphophysiology, life history, and fossil record: exploring key innovations of a diversification sequence Acta Zool. Fenn. 51 2014 80–94 Collier A. Huin J. Nouvelles données sur la faune de mammifères miocènes du bassin de Thenay-Pontlevoy (Loir-et-Cher, France) Bull. Soc. Hist. Nat. Toulouse 113 1–2 1977 219–233 Costeur L. The petrosal bone and inner ear of Micromeryx flourensianus (Artiodactyla, Moschidae) and inferred potential for ruminant phylogenetics Zitteliana B 32 2014 1–16 Costeur L. Mennecart B. Schulz G. Müller B. Prenatal growth stages show the development of the ruminant bony labyrinth and petrosal bone J. Anat. 230 2 2016 347–353 10.1111/joa.12549 Costeur L. Mennecart B. Schulz G. Müller B. Middle ear bones of a mid-gestation ruminant foetus extracted from X-ray computed tomography Proc SPIE 9967 2016 99671Q-2 Costeur L. Schulz G. Müller B. High-resolution X-ray computed tomography to understand ruminant phylogeny Proc. SPIE 9212 2014 921216-1-921216-7 Dubied M. Gilbert C. Deleglise M. Laurens F. Mennecart B. De l’importance de la collection Quatrehomme (Musée La Monnaye, Meung-sur-Loire) dans le paysage paléontologique français Carnets Geol. 17 5 2017 129–138 10.4267/2042/62541 Dubost G. Charron F. Courcoul A. Rodier A. The Chinese water deer, Hydropotes inermis – A fast-growing and productive ruminant Mamm. Biol. 76 2011 190–195 Ekdale E.G. Comparative anatomy of the bony labyrinth (inner ear) of placental mammals Plos One 8 2013 e66624 Fahlbusch V. Säugerreste (Dorcatherium Steneofiber) aus der miozänen Braunkohle von Wackersdorf/Oberpfalz Mitt. Bayer. Staatssaml. Paläont. Hist. Geol. 25 1985 81–94 Gagnaison C. Les assemblages de vertébrés dans deux sites paléontologiques du bassin miocène de Savigné-sur-Lathan/Noyant-sous-le-Lude : La Guimardière et Pelmer (Maine-et-Loire, France) Geodiversitas 35 1 2013 67–103 Gagnaison C. Castillo L. Grugier O. Renou J.-C. Une hémimandibule de Pliopithecus piveteaui dans le Miocène de Contres (41 France) Symbiose nouvelle série 16 2006 26–29 Gagnaison C. Gagnaison J.-C. Hartmann J.-P. Les fossiles de mammifères miocènes de la collection J.-P. Hartmann conservées dans le Musée du Savignéen Symbiose nouvelle série 23 2009 1–16 Gagnaison C. Guevel B. Xerri S. Sicot J.-L. Villeneuve J.-M. Cossard B. La falunière du Tourrelet (Thenay, Loir-et-Cher, France) : nouvelles données sur les vertébrés des sables continentaux du Miocène moyen (Orléanien supérieur : MN5) Rev. Paleobiol. 31 1 2012 219–234 Geist V. Deer of the world their evolution, behaviour, and ecology 1998 Stackpole Books Mechanicsburg 421 p Gentry A.W. Rössner G.E. Heizmann G.E.E.P.J. Suborder Ruminantia Rössner G.E. Heissig K. The Miocene land mammals of Europe 1999 Verlag Dr. Friedrich Pfeil München 225–258 Ginsburg L. Les mammifères terrestres du Miocène moyen des Faluns du bassin de Savigné-sur-Lathan (France) Geodiversitas 23 3 2001 381–394 Ginsburg L. Mornand J. Les restes mammifères des faluns de l’Anjou-Touraine Mém. Soc. Études Sci. Anjou 6 1986 1–18 Grohé C. Tseng Z.J. Lebrun R. Boistel R. Flynn J.J. Bony labyrinth shape variation in extant Carnivora: a case study of Musteloidea J. Anat. 228 3 2015 366–383 10.1111/joa.12421 Hassanin A. Delsuc F. Ropiquet A. Hammer C. Jansen van Vuuren B. Matthee C. Ruiz-Garcia M. Catzeflis F. Areskoug V. Thanh Nguyen T. Coloux A. Pattern and timing of diversification of Cetartiodactyla (Mammalia, Laurasiatheria), as revealed by a com- prehensive analysis of mitochondrial genomes C. R. Biologies 335 2012 32–50 Heydon M. Bolloh P. Mousedeer densities in a tropical rainforest: the impact of selective logging J. Appl. Ecol. 34 1997 484–496 Janis C.M. Lister A. The morphology of the lower fourth premolar as a taxonomic character in the Ruminantia (Mammalia, Artiodactyla), and the systematic position of Triceromeryx J. Palaeontol. 59 2 1985 405–410 Kaup J.J. Description d’ossements fossiles de Mammifères inconnus jusqu’à présent, qui se trouvent au Muséum grand-ducal de Darmstadt, cinquième cahier. J.P. Diehl, Darmstadt 1839 Köhler M. Skeleton and habitat of recent and fossil ruminants Münchner Geowiss. Abh. Geol. Paläont. 25 1993 1–88 Kostopoulos D.S. Sen S. Sen S. Suidae, Tragulidae, Giraffidae, and Bovidae, Late Miocene mammal locality of Küçükçekmece, European Turkey Geodiversitas 38 2 2016 273–298 Mayet L. Étude des mammifères miocènes des sables de l’Orléanais et des faluns de la Touraine Ann. Univ. Lyon 24 1908 1–85 Mennecart B. The Ruminantia (Mammalia, Cetartiodactyla) from the Oligocene to the early Miocene of Western Europe: systematics, palaeoecology and palaeobiogeography Geofocus 32 2012 1–263 Mennecart B. The European ruminants during the “Microbunodon Event” (MP28, latest Oligocene): impact of climate changes and faunal event on the ruminant evolution Plos One 10 2 2015 e0116830 Mennecart B. Becker D. Berger J.-P. Iberomeryx minor (Mammalia, Artiodactyla) from the early Oligocene of Soulce (Canton Jura, NW Switzerland): systematics and palaeodiet Swiss J. Geosci. 104 Suppl. 1 2011 S115–S132 Mennecart B. Costeur L. Shape variation and ontogeny of the ruminant bony labyrinth, an example in Tragulidae J. Anat. 229 3 2016 422–435 10.1111/joa.12487 Mennecart B. Costeur L. A Dorcatherium (Mammalia, Ruminantia, Middle Miocene) petrosal bone and the tragulid ear region J. Vert. Paleontol. 2016 10.1080/02724634.2016.1211665 Mennecart B. Costeur L. 3D models related to the publication: a Dorcatherium (Mammalia, Ruminantia, middle Miocene) petrosal bone and the tragulid ear region MorphoMuseuM 2 2016 10.18563/m3.2.1.e2 (1)-e2 Mennecart B. Métais G. Mosaicomeryx gen. nov., a ruminant mammal from the Oligocene of Europe and the significance of “gelocids” J. Syst. Palaeontol. 13 2015 581–600 Mennecart B. de Perthuis A. Costeur L. 3D models related to the publication: The first French tragulid skull (Mammalia, Ruminantia, Tragulidae) and associated tragulid remains from the Middle Miocene of Contres (Loir-et-Cher, France) MorphoMuseuM 3 2017 10.18563/m3.3.3.e4 (3)-e4 Mennecart B. Rössner G.E. Métais G. DeMiguel D. Schultz G. Müller B. Costeur L. The petrosal and bony labyrinth of Early to Middle Miocene European deer (Mammalia, Cervidae) reveal their phylogeny J. Morphol. 277 2016 1329–1338 Métais G. Vislobokova I. Basal ruminants Prothero D.R. Foss S.C. The evolution of artiodactyls 2007 The Johns Hopkins University Press Baltimore 189–212 Métais G. Chaimanee Y. Jaeger J.-J. Ducrocq S. New remains of primitive ruminants from Thailand: evidence of the early evolution of the Ruminantia in Asia Zool. Scripta 30 2001 231–249 Milne-Edwards A. Recherches anatomiques, zoologiques et paléontologiques sur la famille des chevrotains Ann. Sci. Nat. Paris 5 1864 49–167 Morales J. Sánchez I.M. Quiralte V. Les Tragulidae (Artiodactyla) de Sansan Peigné S. Sen S. Mammifères de Sansan. 203 2012 Mémoires du Muséum national d’Histoire naturelle Paris 225–247 Mottl M. Die Dorcatherien (Zwerghirsche) der Steiermark Mitt. Mus. Bergbau Geol. Techn. Landesmus. Joanneum 22 1961 21–71 Nowak R.M. Walker's mammals of the world 1999 The Johns Hopkins University Press Baltimore and London 1936 p O’Leary M.A. An anatomical and phylogenetic study of the osteology of the petrosal of extant and extinct artiodactylans (Mammalia) and relatives Bull. Am. Mus. Nat. His. 335 2010 1–206 Pfaff C. Matin T. Ruf I. Bony labyrinth morphometry indicates locomotor adaptations in the squirrel-related clade (Rodentia, Mammalia) Proc. R. Soc. B 282 2015 20150744 Rössner G.E. Tragulidae Prothero D.R. Foss S.C. The evolution of artiodactyls 2007 The Johns Hopkins University Press Baltimore 213–220 Rössner G.E. Systematics and palaeoecology of Ruminantia (Artiodactyla, Mammalia) from the Miocene of Sandelzhausen (southern Germany, Northern Alpine Foreland Basin) Paläont. Z. 84 2010 123–162 Rössner G.E. New palaeontological and geological age constraints for the arrival and dispersal of Miocene tragulids (Tragulidae, Artiodactyla) in Europe Zitteliana 89 2017 351–359 Rössner G.E. Heissig K. New records of Dorcatherium guntianum (Tragulidae), stratigraphical framework, and diphyletic origin of Miocene European tragulids Swiss J. Geosci. 106 2013 335–347 Sánchez I.M. Quiralte V. Morales M. Pickford M. A new genus of tragulid ruminant from the early Miocene of Kenya Acta Palaeontol. Pol. 55 2010 177–187 Sánchez I.M. Quiralte V. Rios M. Morales J. Pickford M. First African record of the Miocene Asian mouse-deer Siamotragulus (Mammalia, Ruminantia, Tragulidae): implications for the phylogeny and evolutionary history of the advanced selenodont tragulids J. Syst. Palaeontol. 13 2015 543–556 Seehuber U. Ein Schädelfund von Dorcatherium crassum (Lartet, 1851) aus der Oberen Süßwassermolasse bei Thierhaupten (Landkreis Augsburg, Bayern, Deutschland) Doc. Nat. 196 2 2015 41–61 Smit-van Dort M. Skin, skull and skeleton characters of the mouse deer (Mammalia, Tragulidae), with keys to the species Bull. Zoöl. Mus. Amsterdam 12 5 1989 89–95 Stehlin H.G. Catalogue des ossements de Mammifères tertiaires de la collection Bourgeois à l’école de Pontlevoy (Loir.et.Cher) Bull. Soc. Hist. Nat. Anthr. Loir-et-Cher 18 1925 77–277 Van der Geer A.A.E. Island ruminants and parallel evolution of functional structures Quat. hors-série 2 2005 231–240 Van der Made J. Pre-Pleistocene Land Mammals from Crete Reese D.S. Pleistocene and Holocene Fauna of Crete and Its First Settlers 1996 Prehistory Press Madison 69–79 Vrba E.S. Vaisnys J.R. Gatesy J.E. DeSalle R. Wei K.-Y. Analysis of paedomorphosis using allometric characters: the example of Reduncini (Bovidae, Mammalia) Syst. Biol. 43 1 1994 92–116

Location of the Faluns formation and outline of the Falun Sea (France). The yellow patches correspond to Faluns deposits. The light blue represents the maximum of extension of the Langhian Sea. The white star locates the Contres locality. The figure has been modified from Dubied et al. (2017).

Localisation de la formation des Faluns et contour de la mer des Faluns (France). Les zones jaunes correspondent aux dépôts de faluns. La zone bleutée représente l’extension maximale de la mer Langhienne. L’étoile blanche localise Contres. Cette figure a été modifiée à partir de Dubied et al. (2017).

Dental nomenclature (mainly based on Bärmann & Rössner, 2011). a, upper molars: 1, preprotocrista; 2, protocone; 3, postprotocrista; 4, premetaconulecrista; 5, metaconule; 6, postmetaconulecrista; 7, postmetacrista; 8, metacone; 9, premetacrista; 10, postparacrista; 11, paracone; 12, preparacrista; 13, paracone rib; 14, parastyle; 15, mesostyle; 16, metastyle; 17, cingulum; b, lower molars: 1, preprotocristid; 2, protoconid; 3, internal postprotocristid; 4, external postprotocristid; 5, prehypocristid; 6, hypoconid; 7, posthypocristid; 8, premetacristid; 9, metaconid; 10, internal postmetacristid; 11, external postmetacristid; 12 preentocristid; 13, entoconid; 14, postentocristid; 15 posthypoconulidcristid; 16, hypoconulid; 17, prehypoconulidcristid; 18, ectostylid; 19, anterior cingulid; 20, Tragulus fold; 21, Dorcatherium fold.

Nomenclature dentaire (modifiée d’après Bärmann & Rössner, 2011). a, molaires supérieures : 1, préprotocrista ; 2, protocone ; 3, postprotocrista ; 4, prémetaconulecrista ; 5, métaconule ; 6, postmétaconulecrista ; 7, postmétacrista ; 8, métacone ; 9, prémétacrista ; 10, postparacrista ; 11, paracone ; 12, préparacrista ; 13, côte du paracone ; 14, parastyle ; 15, mésostyle ; 16, métastyle ; 17, cingulum ; b, molaires inférieures: 1, préprotocristide ; 2, protoconide ; 3, postprotocristide interne ; 4, postprotocristide externe ; 5, préhypocristide ; 6, hypoconide ; 7, posthypocristide ; 8, prémétacristide ; 9, métaconide ; 10, postmétacristide interne ; 11, postmétacristide externe ; 12 préentocristide ; 13, entoconide ; 14, postentocristide ; 15 posthypoconulidcristide ; 16, hypoconulide ; 17, préhypoconulidcristide ; 18, ectostylide ; 19, cingulide antérieur ; 20, Tragulus fold ; 21, Dorcatherium fold.

Skull of Dorcatherium crassum from Contres (original specimen AP2014.0715 and digital copy NMB Fa.213.abg). a & c lateral (left), b & d dorsal, e ventral, f posterior view. Abbreviations: boc, basioccipital; bsp, basisphenoid; fr, frontal; jp, jugular process; mx, maxilla; na, nasal; nc, nuchal crest; nt, nuchal tubercule; oc, occipital condyle; or, orbit; pa, parietal; pe, petrosal; pf, postorbital bar of the frontal process; pfs, parietofrontal suture; sc, sagittal crest; sf, supraorbital foramen; sq, squamosal; vo?, vomer?; zy, zygomatic. The scale bar is 5 cm.

Crâne de Dorcatherium crassum de Contres (spécimen original AP2014.0715 et copie digitale NMB Fa.213.abg). 1 Morphologie externe du crâne en vues 1a & c latérale (gauche), 1b & d dorsale, 1e ventrale, 1f postérieure. Abréviations : boc, basioccipital ; bsp, basisphénoïde ; fr, frontal ; jp, processus jugulaire ; mx, maxillaire ; na, nasal ; nc, crête nuchale ; nt, tubercule nuchal ; oc, condyle occipital ; or, orbite ; pa, pariétal ; pe, os pétreux ; pf, barre postorbitale du processus frontal ; pfs, suture pariétofrontale ; sc, crête sagittale ; sf, foramen supraorbital ; sq, squamosal ; vo ?, vomer ? ; zy, zygomatique. La barre d’échelle est de 5 cm.

a right petrosal bone and b left bony labyrinth of Dorcatherium crassum from Contres (original specimen AP2014.0715 and digital copy NMB Fa.213.abg): a1 and b2 ventrolateral view, a2 and b1 dorsomedial view. Scale bare is 1 cm. Abbreviations: aa, asc ampulla; asc, anterior semi-circular canal; ca, cochlear aqueduct; cc, common crus; co, cochlea; es, endolymphatic sac; fc, fenestra cochleae; fv, fenestra vestibuli; la, lsc ampulla; lsc, lateral semi-circular canal; pa, psc ampulla; psc, posterior semi-circular canal; sac, sacculus; sl, secondary bony lamina; ut, utriculus; va, vestibular aqueduct.

Os pétreux droit (a) et labyrinthe osseux gauche (b) du crâne de Dorcatherium crassum de Contres (spécimen original AP2014.0715 et copie digitale NMB Fa.213.abg) : a1 et b2 en vue ventro-latérale et a2 et b1 en vue dorso-médiale. La barre d’échelle est de 1 cm. Abréviations : aa, ampoule asc ; asc, canal semi-circulaire antérieur ; ca, aqueduc cochléaire ; cc, crus commun ; co, cochlée ; es, sac endolymphatique ; fc, fenestra cochleae ; fv, fenestra vestibuli ; la, ampoule lsc ; lsc, canal semi-circulaire latéral ; pa, ampoule psc ; psc, canal semi-circulaire postérieur ; sac, sacculus ; sl, lamina osseuse secondaire ; ut, utriculus ; va, aqueduc vestibulaire.

Dental remains and postcranial bones of Dorcatherium crassum from Contres. 1 Right corpus mandibulae with the tooth row p2–m3 (original specimen AP2014.0129 and cast NMB Fa.215.abg) in 1a labial, 1b occlusal, and 1c lingual views; left Mt III-IV (original specimen AP2015.0128 and cast NMB Fa.216.abg) in 2a plantar, 2b medial, 2c dorsal, 2d proximal, and 2e distal views. The scale bar is 5 cm.

Restes dentaires et os postcrâniens de Dorcatherium crassum de Contres. 1 Corpus mandibulae droit avec la rangée dentaire p2–m3 (spécimen original AP2014.0129 et moulage NMB Fa.215.abg) en vues 1a labiale, 1b occlusale, et 1c linguale; Mt III-IV gauche (spécimen original AP2015.0128 et moulage NMB Fa.216.abg) en vues 2a plantaire, 2b médiale, 2c dorsale, 2d proximale et 2e distale. La barre d’échelle est de 5 cm.

Variation in the p4 in D. crassum from the Auger quarry (Contres, France). a, AP2012.0130 (NMB Fa.214.abg cast); b, AP2014.0129 (NMB Fa.215.abg cast).

Variation de la p4 chez D. crassum de la carrière Auger (Contres, France). a, AP2012.0130 (NMB Fa.214.abg cast); b, AP2014.0129 (NMB Fa.215.abg cast).

Measurements of the bony labyrinth of D. crassum from the Auger quarry (Contres, France) with comparative measurements from the literature (Mennecart & Costeur, 2016b). ASC: anterior semicircular canal; LSC: lateral semicircular canal; PSC: posterior semicircular canal.

Dimensions du labyrinthe osseux de D. crassum de la carrière Auger (Contres, France) et comparaison avec les données de la littérature (Mennecart & Costeur, 2016b). ASC : canal semi-circulaire antérieur ; LSC : canal semi-circulaire latéral ; PSC : canal semi-circulaire postérieur.

Table 1 AP2014.0715 NMB San.15053 Contres MN5 Sansan MN6 Number of cochlear turns 3 3 Volume (mm3) 83.2 67.6 Stapedial ratio 1.62 1.62 Cochlear aspect ratio 0.55 0.63 ASC height (mm) 4.76 4.71 ASC width (mm) 4.54 4.32 PSC height (mm) 4.21 4.15 PSC width (mm) 4.7 4.05 LSC length (mm) 3.93 3.82 LSC width (mm) 3.43 4.18 Angle ASC–PSC 77° 84° Angle ASC–LSC 73° 73° Angle PSC–LSC 80° 85°

Dental measurements (maximal length and width respectively in mm) of D. crassum from the Auger quarry (Contres, France) with comparative material from the literature (Morales et al., 2012, Rössner, 2010 and Seehuber, 2015). Thierhaupten tooth averages are based on left and right dentition of the skull.

Dimensions des dents (longueur et largeur maximales respectivement en mm) de D. crassum de la carrière Auger (Contres, France) et comparaison avec les données de la littérature (Morales et al., 2012, Rössner, 2010 and Seehuber, 2015). Les tailles moyennes pour le spécimen de Thierhaupten sont basées sur les deux rangées dentaires du crâne.

Table 2 AP2012.0130 AP2014.0129 AP2015.0131 Contres MN5 Contres MN5 Contres MN5 p2 – – 11.3 4.4 – – p3 11.8 5.0 11.5 5.1 – – p4 10.6 5.5 11.5 5.8 – – m1 10.5 7.0 10.4 7.5 – – m2 11.4 8.1 12.6 8.5 12.1 8.7 m3 18.7 9.2 19.5 9.0 – 9.2 AP2015.0133 AP2013.0132 Contres MN5 Contres MN5 p2 – – – – p3 12.2 4.5 12.9 4.6 p4 11.1 5.4 – – m1 11.3 7.5 – – m2 – – – – m3 – – – – AP2014.0715 AP2013.135 AP2015.0134 Contres MN5 Contres MN5 Contres MN5 P4 9.1 10.0 9.2 10.6 – – M1 10.6 11.6 11.1 12.0 11.1 12.1 M2 12.1 13.4 12.5 13.8 12.0 13.7 M3 13.0 13.7 13.5 14.3 13.2 14.4 Seehuber, 2015 Morales et al., 2012 Rössner, 2010 (Thierhaupten MN5) (Sansan MN6) (Sandelzhausen MN5) Average Average Average Average Min–max Min–max p2 – – 10.9 4.4 10.6–11.5 4.2–5.5 p3 – – 10.5 4.6 11.3–12.5 4.0–5.2 p4 – – 10.0 5.3 10.8–13.0 5.2–6.0 m1 – – 9.5 7.4 10.5–11.9 6.3–8.5 m2 – – 10.1 8.8 11.0–14.0 7.4–10.0 m3 – – 17.2 8.8 17.5–20.9 7.9–10.2 P2 13.0 6.4 13.2 6.3 13.0 6.6 P3 6.6 8.0 11.3 7.4 11.0 7.5 P4 9.8 10.8 9.1 10.0 10.0 10.0–11.9 M1 10.1 11.6 10.3 11.8 10.0–10.3 11.0–11.1 M2 11.3 13.1 11.9 13.9 11.1–12.3 12.8–14.0 M3 12.5 14.3 13.3 14.7 12.0–12.7 14.1–14.5

Measurements of metatarsal bones (in mm) of D. crassum from the Auger quarry (Contres, France) with comparative measurements from the literature (Morales et al., 2012 and Rössner, 2010). DAP: dorso-palmar diameter; dist: distal; DT: latero-medial diameter; prox: proximal.

Dimensions des métatarses (en mm) de D. crassum de la carrière Auger (Contres, France) et comparaison avec les données de la littérature Morales et al., 2012 and Rössner, 2010). DAP : diamètre dorso-palmaire ; dist : distal ; DT : diamètre latéro-médial ; prox : proximal.

Table 3 Mt III–IV Morales et al., 2012 Rössner, 2010 AP2015.0128 (Sansan Mn6)Average (Sandelzhausen) Length 92.8 96 93.6 DT prox 20.1 – 20.5 DAP prox 18 – – DT dist 25.4 – 24.6 DAP dist 14.5 – 14.7

Summary of the potential heterochronic skull elements and life history (data from Dubost et al., 2011 for living tragulids and Rössner, 2010 for Dorcatherium crassum body weight).

Résumé des éléments crâniens potentiellement hétérochroniques et stratégie de reproduction (données issues de Dubost et al., 2011 concernant les Tragulidae actuels et Rössner, 2010 pour le poids de Dorcatherium crassum).

Table 4 Dorcatherium Hyemoschus Moschiola Tragulus Supraorbital foram ++ – – – Postorbital bar ++ – – – Sagittal crest ++ + – – Nuchal crest ++ + – – Occipital condyles somewhat detached + + – – Nuchal tubercules ++ – – – Jugular processes ++ + – – Gestation time (days) ? 225 150 154–140 Age at weaning (weeks) ? 12 ? 10–11.5 Sexual maturity (months) ? 18 ? 5.5–4.5 Longevity (years) ? 13 ? 15–12 Bodymass (kg) 15 11 2.5 1–7