Introduction

The phylogenetic and biogeographic relationships of fossil European Monitor lizards have recently been revised by Conrad et al. (2009, 2011), and Delfino et al. (2011) focused only on the fossils referred to Varanus. Conrad (2008) and Conrad et al. (2008) only minimally commented on Iberovaranus Hoffstetter, 1969, a monotypic genus that was erected on the basis of a single trunk vertebra from the Miocene of Can Mas, close to El Papiol, about 15 km from Barcelona (Catalonia, Spain). Hoffstetter (1969) considered Iberovaranus to be the most primitive member of the Varaninae, but owing to the poor quality of the material, Clos (1995) and Conrad et al. (2011) regarded Iberovaranus as too fragmentary to be included in phylogenetic analyses. Nonetheless, other authors identified the morphological characters of Iberovaranus from other localities, sometimes with doubts concerning the specific identification. For example, Antunes and Rage (1974) and Antunes and Mein (1981) referred to Iberovaranus a few isolated vertebrae from three early—middle Miocene localities of Portugal. Böhme and Ilg (2003) listed about 10 more localities from Spain and Germany with possible remains of Iberovaranus, with an age span from Burdigalian (MN3) to Langhian (MN6). Here we re-examine the holotype of Iberovaranusm and another specimen from the same locality, in order to consider its validity.

Institutional abbreviations.—ICP, Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain; IPS, collections of the ICP (formerly Institut de Paleontologia de Sabadell); ZFMK, Zoologisches Forschungsmuseum Koenig, Bonn, Germany.

Other abbreviations.—CMA, Can Mas, El Papiol, Barcelona, Spain.

Geological and geographical setting

CMA is a loosely-defined paleontological site, which includes Miocene levels of reddish to brownish clays from several quarries that were situated close to the house of Can Mas, about 1 km NW from the town of El Papiol (Crusafont et al. 1955). This locality, which should not be confused with the homonymous, Vallesian locality of “Can Mas”, in the area of els Hostalets de Pierola (Crusafont and Truyols 1954)—corresponds to the early Miocene Lower Continental Complex of the Vallès-Penedès Basin; it is mainly composed of reddish terrigenous deposits of varied lithology, deposited in the context of alluvial fan systems (Agustí et al. 1985; Casanovas-Vilar et al. 2011b). An MN4 age (i.e., circa 17–16 Ma; Agustí et al. 2001) has been customarily given to CMA (Agustí et al. 1985; Fortelius 2011), which is consistent with the presence of Gomphotherium angustidens (cited by Crusafont et al. 1955), although this taxon does not exclude an earlier, MN3 age (Agustí et al. 2001; Casanovas-Vilar et al. 2011a). Further refinement is precluded by the lack of associated rodents. An attribution to MN4 would also be supported by the record of Eotragus reported by Fortelius (2011). However, an inspection of the available macromammalian fossils from CMA housed at the ICP yielded no remains of the bovid Eotragus. Interestingly a single mandibular fragment of the paleomerycid Lagomeryx (DMA, personal observations), was located, but does not discriminate between MN3 and MN4.

Systematic palaeontology

Fig. 1.

Varanid lizard Varanus sp. from Can Mas, early Miocene, IPS 58437; holotype of Iberovaranus catalaunicus Hoffstetter, 1969, a taxon here considered as taxonomically non valid. Trunk vertebra in dorsal (A), ventral (B), anterior (C), posterior (D), and left lateral (E) views.

Fig. 2.

Varanid lizard Varanus sp. from Can Mas, early Miocene, IPS 31701. Cervical vertebra dorsal (A), ventral (B), anterior (C), posterior (D), and left lateral (E) views.

IPS 31701 is a partially preserved posterior cervical vertebra from CMA previously catalogued as “cf. Testudo sp.?” (Fig. 2). The neural spine, the left parapophysis, and the preand postzygapophyses are broken off, but the ventral surface of the centrum is well preserved. The centrum, 9.5mmlong, is slightly convex transversely and not as triangular in shape as that of IPS 58437. A hint of a median keel is present in its anterior sector. The centrum has a distinct, obtuse precondylar constriction (width at the constriction 4.9 mm; maximum width of the cotyle 6.2 mm; ratio 0.78), which involves a significant portion of the lateral side of the centrum. The cotyle and condyle are markedly dorsoventrally depressed; although the dorsal rim of the cotyle, as preserved, did not overhang the preceding condyle, this is likely an artefact of erosion. The ventral edge of the condyle is close to the posterior edge of the centrum (so that only a little portion of the condyle is visible in ventral view). The synapophyses are massive, nearly vertical and strikingly laterally projected, so that the vertebra is 12.3 mm wide at their level. They are ventrally linked to the edge of the cotyle by a sub-horizontal ridge (“ventral parapophyseal lamina” in Houssaye et al. 2011). The diapophysis and parapophysis are well defined and slightly separated by a posterior concavity; the former is larger than the latter. Between the parapophyses and the rim of the cotyle there is shallow concavity posteromedially marked by a step.

Discussion

After the remarkable but partial works by Mertens (1942, 1950, 1959), knowledge of the osteology of genus Varanus has been advanced thanks to the identification of a few hundred phylogenetically informative characters summarized and updated by Conrad et al. (2011). Although the fine morphology of the 73 species currently ascribed to Varanus (Uetz et al. 2011) is still imprecisely known and the intraand interspecific variation of several characters largely underexplored, it is now clear that the degree of precondylar constriction can significantly vary among species. The constriction is always present in Varanus, yet can vary from weakly expressed (state 1 of character 233) to so well developed that it is less than 80% of the maximum condylar diameter and “a right or acute angle is formed between the condyle and the centrum just anterior to the posterolateral part of the condyle (state 2 of character 233)” (Conrad et al. 2011: 267).

According to Hoffstetter (1969), Iberovaranus differs from Varanus in having a more elongate neural arch, a narrower condyle, and a less pronounced precondylar constriction. These presumed differences are partly explained by the preservation of the type vertebra (condyle, pre and postzygapophyses are somewhat eroded) and partly to interspecific variability within Varanus. In particular, the vertebrae of some extant species, such as Varanus griseus (Daudin, 1803) (Fig. 3A),match well with the trunk vertebra IPS 58437. The precondylar constriction of the latter is of 84%, corresponding to state 1 of character 233 (Conrad et al. 2011), but this result is considerably altered by the erosion of the condyle. In fact, the constriction of posterior cervical vertebra IPS 31701, which has a perfectly preserved condyle, is 78%, corresponding to state 2, indicating therefore that this could likely be themorphology and state of the vertebrae of the CMAmonitor lizard. The average precondylar constriction of the trunk vertebrae of V. griseus ZFMK 14883 and ZFMK 53533 is about 76%, corresponding again to state 2, even if the general shape of the centrum is quite different from that of other congeneric species, such as Varanus salvator (Laurenti, 1768), with the same status (see Fig. 3B).

Fig. 3.

Trunk vertebrae of extant varanid lizards in ventral view. A. Varanus griseus (Daudin, 1803), ZFMK 53533. B. Varanus salvator (Laurenti, 1768), ZFMK 90471. Note the different development of the precondylar constriction (indicated by the arrows) and the proportions of the centrum and the condyle.

Finally, it is worth mentioning that, according to Molnar (2004), the vertebrae of Iberovaranus catalaunicus resemble those of helodermatids because of the slight development of the precondylar constriction. Although it is true that the condition of IPS 58437 could be considered intermediate between that shown by, for example, V. salvator and helodermatids, the absence of any constriction in the latter (Conrad et al. 2011; state 0 of character 233), among other characters, excludes referral of IPS 58437 to these taxa.

A further character supporting the inclusion of the holotype of Iberovaranus, IPS 58437, in Varanus is the presence of striations of the neural arch, which are considered, along with the precondylar constriction, an apomorphy of Varanus (Smith et al. 2008). Such striations are not present in all Varanus species but “have arisen on the Varanus stem and are retained as plesiomorphy in many crown Varanus, especially the African forms” (Smith et al. 2008: 911).

Conclusions

The morphology of the type vertebra of Iberovaranus catalaunicus Hoffstetter, 1969, IPS 58437, as well as that of the only other varanoid vertebra from the same locality, IPS 31701, is clearly contained within the variability of extant Varanus. The nominal taxon Iberovaranus Hoffstetter, 1969 must therefore be considered a subjective junior synonym of Varanus Merrem, 1820. The taxonomic validity of the nominal taxon Iberovaranus catalaunicus Hoffstetter, 1969, the type and only species of this genus, is not supported by the characters already mentioned for the genus. Given that it is available in terms of nomenclature, but from a taxonomic viewpoint it cannot be either validated or synonymized with other Varanus species on the basis of currently available material, it is here deemed a nomen dubium (see Mones 1989, for a review of this concept). The species binomen Varanus catalaunicus (Hoffstetter, 1969) comb. nov. could potentially be used in the future if a new comparative diagnosis becomes possible on the basis of new skeletal elements providing significant morphological information (e.g., skull elements; see Conrad et al. 2009; Delfino et al. 2011).

If the referral to MN3 of the CMA vertebrate assemblage is confirmed, the Varanus remains here described represent one of the earliest evidences of the dispersal of this genus into Europe, which is otherwise known with confidence from the MN4b of Béon 1 (Rage and Bailon 2005) and possibly from a few MN3 Spanish localities such as San Roque 4A (Varanus sp.) and Agramon and Ateca 1 (aff. Iberovaranus sp. and Iberovaranus cf. I. catalaunicus; Böhme and Ilg 2003).