
A pressing issue in historical biogeography for clades with long evolutionary histories and intercontinental distributions is the ever-changing position of tectonic plates. Over time, this shifting alters the proximity between areas, a factor that very few models can easily account. In addition, the absence of fossils from most molecular phylogenies often erases biogeographic signals not retained in trees of extant taxa. Here, we adapt the paleogeographic model from Landis (2017), a potentially powerful tool for ancestral-range estimation at global scales through time. This model creates “communicating classes” of areas by assigning discrete geographic units to adjacency matrices that change across multiple time slices in the Phanerozoic to model continental drift. We apply this algorithm to three existing total-evidence datasets incorporating extinct and extant tips from the order Squamata. Our results corroborate the origin of Squamata on the Eurasian continent – specifically Europe and northeastern Asia – although phylogenies sampling more Jurassic squamate lineages showed higher support for a purely European origin. Eurasia continued to be a major source of diversification throughout the Mesozoic, with dispersals into North America in the Late Jurassic and South America in the mid-Cretaceous. For Serpentes, the ancestral ranges were unclear and inconsistent across the phylogenetic hypotheses, likely influenced by the disparate and incomplete sampling in the three phylogenies.
Biogeography, Squamata, snakes, lizards, tectonics, paleobiogeography, fossil