The fossil record of Ceratophrys Wied-Neuwied (Anura: Ceratophryidae): a revision and update of fossil South American horned frogs Author Nicoli, Laura text Zootaxa 2019 2019-08-21 4658 1 37 68 journal article 25962 10.11646/zootaxa.4658.1.2 ada41293-20e6-4824-a8c4-f7e8eb6d16b4 1175-5326 3373075 16EDCB6E-49D1-4214-AEB3-203C19CA56A0 Ceratophrys ensenadensis Rusconi 1932 Ceratophrys ensenadensis is based on three specimens from Olivos ( Fig. 1 , locality 10), province of Buenos Aires , Argentina . In this area, as in other regions of the Rio de la Plata, a series of deposits attributed to the Ensenada Formation, known as “toscas del Río de la Plata”, were exposed and provided many fossils ( Soibelzon et al. 2008 ). Given the growth of the city of Buenos Aires , and the reclamation of land along the river, these deposits are now confined to small areas. The “toscas of Rio de la Plata” are thought to have been deposited during the early and middle Pleistocene ( Soibelzon et al. 2008 ). FIGURE 2. Ceratophrys ameghinorum compared with selected living Ceratophrys . A, D, G, J, Ceratophrys aurita (MLP 1280); B, E, H, K, C. ameghinorum (MACN 14318); C, F, I, L, C. cornuta (MZUSP 151382). A–C, dorsal views; D–F, ventral views; G–I, lateral views; J–L, ventrolateral view. Abbreviations: ap, premaxillary alary process; lc, lateral crest; m, maxilla; mp, maxillary process of nasal; mpp, maxillary process of premaxilla; par, pterygoid anterior ramus; pl, neopalatine; pmr, pterygoid medial ramus; pof, postorbital fenestra; ppa, discrete pars palatina absent; ppr, pterygoid posterior ramus; pt, pterygoid; slw, squamosal lateral world; sop, squamosal otic plate; v, vomer; vp, ventral process of pterygoid. Scale bar is 5 mm. Rusconi briefly described the fossil remains, but he did not provide photographs or illustrations (Rusconi 1932). The author identified the material as part of the Henning personal collection. This collection is now housed at the Museo Universitario Florentino y Carlos Ameghino (MUA) of Rosario (province of Santa Fe , Argentina ). No remains of Ceratophrys ensenadensis or other horned frogs were found in this collection (Silvia Cornero, MUA director, pers.comm.). The remains of two fossil horned frogs (PVL 699, 767) accessioned as C. ensenadensis were found in the Collection of Vertebrate Paleontology of the Instituto Miguel Lillo of San Miguel de Tucuman (province of Tucuman , Argentina ; Fig. 3A, B, G, H ). These specimens were part of the personal collection of Carlos Rusconi (414 and 413, respectively). Based on the description and measurements of these specimens provided by Rusconi (1932), they are clearly the holotype (PVL 767; Fig. 3A, B ) and one of the paratypes ( Fig. 3G, H ) of C. ensenaden- sis . An illustration, which probably corresponds to PVL 699, was published and attributed to C. ensenadensis by Agnolin (2005: fig.1), but the collection number cited (PVL 616) is wrong. The third specimen originally mentioned by Rusconi remains lost. Osteological features. Specimen PVL 767 is a partially preserved skull ( Fig. 3A, B ). It is too incomplete to estimate its proportions. The cranial elements that are preserved generally resemble those of extant Ceratophrys . The squamosal otic plate is subquadrangular, and lacking a concave dorsal surface or lateral wall. Its posterior margin reaches the level of the occipital condyles, and do not project dorsally from the skull roof. The premaxilla was not preserved. The vomers seem to bear teeth. PVL 699 consists of the posterior region of a skull that articulates with the vertebral column ( Fig. 3G, H ). Morphologically, it resembles the same region of extant Ceratophrys . The part of the squamosal otic plate that is preserved is subquadrangular; the morphology of the posterolateral region cannot be assessed because it seems to be broken.An incomplete, although extensive, dorsal shield formed by several distinct and irregular plates is preserved over the vertebral column (extending over Presacrals I–VII). Remarks. Both specimens have all of the diagnostic features of Ceratophryidae and Ceratophrys that can be evaluated ( PVL 767: exostosis, parieto-squamosal arch, expanded squamosal otic ramus overlapping prootic, absence of pars palatina of maxilla [ Ceratophryidae ]; nasals with robust, bar-shaped maxillary processes; postorbital fenestra [ Ceratophrys ]. PVL 669: exostosis, parieto-squamosal arch, expanded otic ramus of squamosal overlapping prootic, transverse processes of the most anterior presacral vertebra distinctly expanded and long [ Ceratophryidae ]; postorbital fenestra [ Ceratophrys ]). Thus, the generic assignment of both remains seems to be well supported. Rusconi’s (1932) definition of Ceratophrys ensenadensis was fundamentally based on alleged differences between the fossils and C. ornata . He reported that: the nasals are more robust and expanded anteriorly in C. ensenadensis than in C. ornata ; the dermal ornamentation is high and tuberculate in C. ensenadensis in contrast to a reticulate pattern in C. ornata ; and the dorsal shield is more expanded in C. ensenadensis . Rusconi (1932) also mentioned differences in the posterior region of skull, but it is difficult to determine what, exactly, he described. For example, he refers to the “jugal” bone, but this element is absent in anurans; possibly he was referring to the squamosal or quadratojugal. Likewise, he references some osseous protuberances dorsal to the occipital condyles. Previous authors have expressed that this fossil species might be invalid ( Báez & Gasparini 1977 ; Perí 1993a ). The morphology of the specimens thought to be the holotype ( PVL 767) and one of the paratypes ( PVL 669) of Ceratophrys ensenadensis is subsumed by the intraspecific variation observed in the clade [ C. cranwelli – C. ornata ] ( Fig. 3 ).The proportions of the skull (including the partially preserved nasals) and of the dorsal shield resemble those observed in several specimens of the living species. The dermal ornamentation is also similar to that of extant membersof this clade ( Fig. 3 ). The morphology of the dermal ornamentation has been a recurrent theme in taxonomic studies of Ceratophrys (e.g., Ameghino 1899 , Scanferla & Agnolín 2015 ). However, dermal sculpturing is extremely variable intraspecifically, ontogenetically, and even topographically in the same individual (Nicoli 2017; Fig. 3 ). The tuberculation described by Rusconi does not differ from that observed in some individuals of C. cranwelli and C. ornata ( Fig. 3 ). Thus, there seems to be insufficient evidence to support the specific status of C.ensenadensis . The holotype of Ceratophrys ensenadensis ( PVL 767) possesses several presumably pleisiomorphic features of Ceratophrys (e.g., subquadrangular squamosal otic plate lacking a dorsal projection and a concave dorsal surface or lateral wall, and extending posteriorly to the level of the occipital condyles). Moreover, PVL 699 is too fragmentary to support its reference to a less-inclusive taxon within Ceratophrys . Although the presence of a dorsal shield may be a synapomorphy of the southeastern clade, its presence in C. ameghinorum , which has an unresolved phylogenetic position within Ceratophrys , precludes inclusion of PVL 699 in that clade. Additionally, there is no evidence that both of these fossils belong to the same species. Features of both fossil specimens support their referral to Ceratophrys , but not their assignment to a less-inclusive taxon. The present study concludes that Ceratophrys ensenadensis is not a valid species. Furthermore, its type series cannot be unequivocally assigned to any other species, and thus C. ensenadensis cannot be placed in the synonymy of another species. Because C.ensenadensis has a designated holotype , the name cannot be discarded without further evidence and, thus, is considered species inquirenda .