Not knowing what you don't know: Geography and dispersal in Lepidoptera biogeography

An inferred trans-oceanic dispersal of Painted Lady butterflies was recently presented as evidence for chance dispersal being responsible for allopatric distributions of taxa in support of the belief that the biogeographic distributions and phylogenetic histories of many insects could not otherwise be explained. The six studies cited as providing evidence for chance dispersal origins are shown to share assumptions about molecular divergence estimates, chance dispersal, centers of origin, and ancestral areas computations that render their findings highly problematic, if not invalid. The approach used in these studies exemplifies the absence of geographic distribution as a direct subject of comparative biogeographic analysis. The analytical significance of this absence is illustrated for Arhuaco butterflies in Central America where the origin of a disjunct species was attributed to Pleistocene chance dispersal based on an assumption of recent divergence. Comparison of Arhuaco distribution with other taxa suggests its location and differentiation was affected by geological displacement between the Romeral and Polochic-Motagua fault zones in late Mesozoic time. This tectonic context is made visible when directly comparing the distributions of different taxa, but lost to view when subordinated to a priori area units in ancestral-area methods.