The Amazon River as a dispersal barrier to passerine birds: effects of river width, habitat and taxonomy

Aim To evaluate the relative effectiveness of the lower and upper sections, respectively, of the Amazon River as a barrier to bird distribution, and to evaluate ecological and taxonomic factors affecting the efficacy of the river barrier. Location Amazon River of South America between its confluence with the Napo River in the west and its delta in the east. Methods Using published distribution maps for 448 species of passerine birds occurring along the Amazon River, we evaluated whether each was distributed along one bank only (river presumed to be a barrier) or both banks (no barrier) to test the predictions that the river was more effective as a dispersal barrier: (1) along the lower, wider portion of the river than the upper, narrower portion; (2) for species inhabiting forests than open country; (3) for species inhabiting forest understorey than forest canopy; (4) for species restricted to terra firme (never inundated upland forest) than those not restricted to terra firme and (5) for certain taxonomic groups. Results Our analyses demonstrated that the Amazon River was most effective as a dispersal barrier along its lower portion and for species restricted to forests and terra firme. However, the river was not significantly more of a barrier for species inhabiting forest understorey than forest canopy. The river was most significant as a barrier to dispersal for the antbirds (Thamnophilidae) and was less significant as a barrier to species belonging to several large families including woodcreepers (Dendrocolaptidae), ovenbirds (Furnariidae), flycatchers (Tyrannidae), cotingids (Cotingidae), tanagers (Thraupidae), seed-eating finches (Emberizidae) and blackbirds (Icteridae). Main conclusions The robust widths of Amazonian rivers are widely considered to represent impediments to dispersal and gene flow for many taxa of birds and other animals, and may have represented agents of vicariance in the diversification of species. Our study reaffirms the effectiveness of the lower Amazon River as a current barrier to bird dispersal for forest birds and provides new insights into the effects of habitat and taxonomy on the efficacy of the river barrier. Although supportive of several predictions of the river hypothesis of biological diversification, our study is limited in addressing the historical impact of river barriers as agents of vicariance in the process of diversification.