A comparison of a discovery-based and an event-based method of historical biogeography

Aim The event-based method Dispersal-Vicariance Analysis (DIVA) is compared with the discovery-based method Brooks Parsimony analysis (BPA). Location South-western USA, Mexico and northern Central America. Methods Results of DIVA of phylogenetic trees for six clades of birds inhab iting seven areas in the south-western US, Mexico and northern Central Amer ica are compared with those of BPA for the same data set. Results Both approaches identify the same vicariant elements but differ in the way they treat dispersal. DIVA places such elements in one general 'dis persal' category, while BPA identifies different forms of dispersal, includ ing peripheral isolates speciation (speciation by dispersal), post-speciati on dispersal, non-response to a vicariance event, secondary contact between congeners (and the potential for reinforcement completing speciation) and potential extinction resulting from competition between a resident and a co lonizing congener. Main conclusions BPA is more sensitive than DIVA with respect to the differ ent possible manifestations of geographical dispersal. Despite substantial dispersal, avian communities in these areas manifest substantial historical structuring.