Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Biological Sciences

First Advisor

James V. Remsen, Jr


Ongoing debate about the evolutionary history of the Amazonian avifauna suffers from a shortage of data. Especially needed are thorough geographic sampling, sensitive markers of genetic diversity, and historical data in the form of explicit phylogenies. I examined patterns of geographic variation in three widespread species complexes of nearly identical Amazonian flycatchers in the genus Hemitriccus, based on spectrographic analysis of vocalizations and phylogenetic analysis of DNA sequences (1026 base pairs of cytochrome b from 44 individuals) from throughout Amazonia. Although members of the three complexes are extremely difficult to distinguish morphologically and have long been confused by taxonomists, they represent three well-defined clades separated by a remarkable 10% sequence divergence. Within clades, vocally defined populations correspond closely to genetic units and more accurately than does current taxonomy. Area cladograms of the three species complexes are not strictly concordant and show five-fold differences in inter-regional sequence divergences, suggesting either different rates of evolution or different ages of vicariance events. Two taxa, the H. zosterops and H. inornatus complexes, show similar, but not identical, patterns of a basal north-south split along the current path of the Amazon River and a more recent east-west split within the northern population, consistent with the possibility of a single set of vicariance events affecting both taxa similarly and simultaneously. Nevertheless, within-population genetic variability differs dramatically between the two groups, suggesting unique histories or life-history traits. The third complex, H. minor , shows a different geographic pattern of distribution and genetic divergence. Presently, no single hypothesis proposed to explain patterns of Amazonian biogeography adequately or uniquely accounts for the patterns found in this study. A null hypothesis is proposed, in which current environmental conditions are sufficient to explain existing patterns without invoking particular historical changes. Based on the results of this study, the null hypothesis cannot be rejected. More such studies on many more taxa will be necessary before evolutionary patterns are sufficiently well described to differentially implicate specific historical processes.