THE FIRST SOUTH AMERICAN FOSSIL RECORD OF SHIELD-TAILED SNAKES (SERPENTES, UROPELTIDAE)

AGUSTÍN SCANFERLA; JUDITH BABOT; DANIEL GARCÍA-LÓPEZ

Villa de Leyva

Congreso; VI Congreso Latinoamericano de Paleontología de Vertebrados; 2018

The family Uropeltidae constitutes a small clade of extant snakes endemic to the moistforests of peninsular India and Sri Lanka. All species of this clade of basal alethinophidianare highly specialized fossorial forms, with extensive anatomical traits for this mode of lifeand a singular mode of locomotion. Particularly, the braincase is characterized by fusion ofbony elements (except the genus Melanophidium), sizable otic capsules, and a roundedoccipital condyle. Up to now, uropeltid snakes have no fossil record. Here, we present thefirst fossil specimen assigned to this clade based on a small incomplete braincase recentlydiscovered in Late Eocene levels of Geste Formation in the Puna of Catamarca province,Argentina. Contrary with the widespread condition present in snakes, in the fossilspecimen the braincase bones are fused. As in miniaturized fossorial squamates, the oticcapsules are enormous, with a large collumelar footplate closing the fenestra ovalis. Theoccipital condyle is rounded (i.e. fovea dentis absent), a condition only present inuropeltids among squamate reptiles. CT reconstructions allow the recognition of afenestra pseudorotunda that communicates the vestibular cavity with the juxtastapedialrecess independently to the recessus scalae tympani. Remarkably, this arrangement isonly present in uropeltids and the rest of basal alethinophidians commonly labelled as“anilioids” (Anilius, Cylindrophis, Anomochilus). As in most uropeltids, the foramen forthe facial nerve converges with the posterior opening of the Vidian canal. Taken together,these traits strongly support this systematic allocation. This unforeseen fossil discovery inSouth America represents a biogeographic challenge due to the restricted Asiatic extantdistribution of uropeltid snakes and the unique paleogeographic history of Indiansubcontinent. This new fossil record and the present distribution in a formerlyGondwanan terrain suggest that the origin of uropeltids occurred in this ancientsupercontinent, thus becoming extant uropeltid species into Gondwanan relicts. Availablecalibrated timetrees and relaxed molecular clocks estimate the split between uropeltidsand the rest of alethinophidian snakes 100-80 MA. Given that Gondwanan break-uphappened mostly before this time interval, the presence of an uropeltid in the Eocene ofSouth America and the extant distribution of this clade in India must have explained viatransoceanic and/or island-hopping dispersion between West Gondwana (South Americaand Africa) and East Gondwana (Antarctica, India, Madagascar and Australia). There is agrowing body of information that suggests a bidirectional faunistic and floristicinterchange between Africa and South America during the Campanian-Eocene time span(80-40 MA) through an island arch established in the Atlantic Ocean (Rio Grande Riseand Walvis Ridge). Furthermore, a dispersal of several vertebrate taxa with Gondwanan affinities to the Indian subcontinent from Africa along the southern margins of theNeotethys around 54 MA ago was proposed recently. In this context, the dispersionbetween India and South America through Africa during a long period between the LateCretaceous and Paleogene seem to be a plausible biogeographic scenario to explain thegeographic distribution of shield-tailed snakes.