But they move! Vicariance and dispersal in southern South America: Using two methods to reconstruct the biogeography of a clade of lizards endemic to South America.

Abstract

This study aims to identify events that modeled the historical biogeography of Phymaturus, using three methodologies: Spatial Analysis of Vicariance (VIP), Statistical Dispersal-Vicariance Analysis (S-DIVA), and Bayesian Binary Method MCMC (BBM). In order to assign areas for the Dispersal-Vicariance and the BBM analyses, we preferred not to use predefined areas, but to identify areas defined via an endemism analysis of Phymaturus species. The analyses were conducted using the same basic topology, which we obtained by constructing a metatree with two recent phylogenies, both morphology and molecular-based. This topology was also used to obtain time divergence estimates in BEAST, using more outgroups than for the metatree in order to get more accurate estimates. The S-DIVA analysis based on the metatree found 25 vicariance events, 20 dispersals and two extinctions; the S-DIVA analysis based on the BEAST tree yielded 30 vicariance events, 42 dispersal events and five extinctions, and the BBM analysis yielded 63 dispersal events, 28 vicariance events and 1 extinction event. According to the metatree analysis, the ancestral area for Phymaturus covers northern Payunia and southern Central Monte. A vicariant event fragmented the ancestral distribution of the genus, resulting in northern Payunia and southern Central Monte as ancestral area for the P. palluma group, and southern Payunia for the P. patagonicus group. The analysis based on the BEAST tree showed a more complex reconstruction, with several dispersal and extinction events in the ancestral node. The Spatial Analysis of Vicariance identified 41 disjunct sister nodes and removed 10 nodes. The barrier that separates the P. palluma group from the P. patagonicus group is roughly congruent with the southern limit of the P. palluma group. The ancestral range for the genus occupies a central position relative to the distribution of the group, which implies that the species must have migrated to the north (P. palluma group) and to the south (P. patagonicus group). To answer questions related to the specific timing of the events, a molecular clock for Phymaturus was obtained, using a Liolaemus fossil for calibration. The present contribution provides a hypothetical framework for the events that modeled the distribution of Phymaturus.