Radiation in the genus Amazilia: A comparative approach to understanding the diversification of hummingbirds.

Abstract

Diversity patterns of morphology and behavior are documented in hummingbirds. I perform a phylogenetic analysis of Amazilia, the largest genus in Trochilidae, and putative sister taxa to test whether the unusual divergence of this genus is an artifact (Chapter 1). Phylogenetic analyses resolve several clusters and show that Amazilia is not monophyletic. Then I narrate the discovery of bill serrations among 28 hummingbird genera (Chapter 2), and hypothesized that these structures aid in nectar-robbing of long-tubed corollas. I test this hypothesis and found that the association between nectar-robbing and bill serrations evolved several times independently within the family (Chapter 3). While the analysis suggests a strong relationship, it does not approach statistical significance. Two competing hypotheses for the evolutionary origin of hummingbirds with bill serrations, monophyletic or polyphyletic, were tested as well. My results show that hummingbirds with bill serrations are polyphyletic in origin. However, assuming a monophyletic origin for hummingbirds with bill serrations and placing genera according to traditional classifications reduced the treelength more steps than placing genera according to the hypothesis of a polyphyletic origin. Chapter 4 describes an aviary experiment conducted to assess differences in color discrimination and learning abilities among resident and migrant hummingbirds. I found that switching was the most efficient behavioral response to changes in the aviary, but some individuals gave up quickly or consistently visited the same color. Species differences in switching are interpreted as: (1) intrinsic differences in the tendency to switch, (2) a differential sensitivity to different colors, (3) learning differences among species, and (4) differences in experience among individuals. Lastly, I discuss several hypotheses to explain some of the patterns of size and plumage variation observed in hummingbirds (Chapter 5). I found that the degree of sexual dimorphism in hummingbirds varies depending upon the mating system, display type, plumage iridescence, and tail shape. Traditionally, sexual size dimorphism has been correlated with mating systems. However, lekking was not correlated with size or plumage dimorphic when phylogenetic history was considered. I interpret the observed patterns, morphological differences between the sexes, as an indirect consequence of sexual selection.