Trait-dependent biogeography: model-based inference of dispersal and distribution patterns of Indo-Pacific trap-jaw ants (Hymenoptera: Formicidae: Odontomachus)

Abstract

Dispersal is influenced by ecology, but many popular biogeographical methods do not consider ecological variation among lineages. Here we use a novel trait-dependent dispersal model to infer the historical biogeography of Indo-Pacific trap-jaw ants (Formicidae: Odontomachus). Our working hypothesis is that macroevolutionary dispersal across archipelagos is influenced by habitat preferences, categorized here simply as undisturbed forests or open/disturbed habitats. Based on a multi-locus, fossil-calibrated phylogeny and the new trait-dependent dispersal model implemented in the R package BioGeoBEARS, we found strong evidence that habitat preference shifts from undisturbed forest to open/disturbed habitats increase dispersal rate. This approach allowed us to expand on E.O. Wilson's seminal work, "The Nature of the Taxon Cycle in the Melanesian Ant Fauna". The Taxon cycle is a non-equilibrium island biogeography model that narrates the tight links among ecology, adaptation, dispersal, and speciation. In line with predictions of the taxon cycle model, transition rates to the forest interior state were significantly higher than to open/disturbed habitat in trap-jaw ants. The phylogenetic predictions outlined in this study can be used in future work to evaluate the relative weights of neutral (e.g., geographical distance and area) and non-neutral processes (trait-dependent, macroevolutionary dispersal) in historical biogeography and community ecology at phylogenetic scale.