Description
Plumage coloration in birds is a phenotypic trait that is important to many aspects avian biology, and has been a subject of study for centuries. It provides a quantifiable trait that is shaped by both natural and sexual selection. Sexual dichromatism, or differences in coloration between sexes, is a result of the accumulation of differential selective pressures on males and females. Pigments must accumulate in order to provide bright coloration, and structural colors must develop in a specific way, making them an honest indicator of an individuals' condition. Thus, sexual selection could drive an individual to have brighter, more colorful plumage. However, bright coloration could also make an individual more susceptible to predation, so natural selection could drive the evolution of dull plumage that blends into the environment. For centuries scientists have debated whether sexual dichromatism, with bright males and dull females, is the result of sexual selection on male plumage, or natural selection on female plumage. Most previous studies of the evolution of sexual dichromatism have classified plumage based on the human visual perspective. However, as the knowledge of differences in the avian and human visual capabilities has increased, it has become apparent that this is not appropriate. Humans have trichromatic visual systems, while birds have tetrachromatic visual systems, with the fourth cone cell sensitive to the ultraviolet (UV) part of the electromagnetic spectrum, increasing their perceivable range to 300 _ 700 nm, while humans can only see between 400 _ 700 nm. In addition, they have oil droplets attached to each of their cone cells that act as long-pass cut-off filters, and enhance birds' discriminatory capabilities in the part of the spectrum that humans can detect. The first objective of this study is to assess the extent of UV coloration in avian plumage, and to see the extent of sexual dichromatism as perceived by the avian visual system in a group of birds called the tanagers and cardinals. Individuals from 401 species were measured using reflectance spectrometry. UV reflectance was found to be common in the patches measured, and all but three species reflected at least 5% of the light in the UV range of at least one of their plumage patches. Almost half (48%) of all species reflected greater than 20% of light in the UV range in at least one patch. Interestingly, more females reflected UV light than males at the 5% and 10-20% levels, but more males showed an average reflectance greater than 20%. High UV reflectance, including 81 out of the 100 patches that were found to be primarily UV colored, belonged to species in three clades, indicating there is some phylogenetic signal to UV coloration. In addition, sexual dichromatism was found to be much more widespread in the group than previously thought. Form a human visual perspective, about half of the species of tanagers and cardinals are sexually dichromatic, but from an avian visual perspective, 97.3% of species are shown to be dichromatic. Like UV reflectance, higher levels of sexual dichromatism were found in some clades than others, indicating phylogenetic signal. The levels of UV reflectance in the tanagers and cardinals were found to be comparable to those found in birds as a whole, and the levels of sexual dichromatism were found to be comparable to those found in passerines, a clade of about half of all bird species. These findings suggest that the variation in both plumage and life history traits present in tanagers and cardinals makes it an appropriate group for studies of behavior that could be applicable to birds as a whole. The second objective of the study was to see if, in the tanagers, it is changes in male or female plumage that are driving the evolution of sexual dichromatism. By quantifying male and female plumage patterning using the avian tetrahedral color space, I was able to look for correlations of male and female plumage with sexual dichromatism, as determined by the avian visual system. I found that when the whole tanager clade was analyzed, there were eight significant male plumage correlations, and two significant female correlations. Thus, male plumage is more often correlated with changes in sexual dichromatism than female plumage. However, these findings broke down when seven tanager subclades were analyzed individually. In four of these clades, there were more male than female correlations (although some only had a difference of one), in two of the clades there were more female than male correlations, and in one clade there were equivalent male and female correlations. This suggests that in some parts of the tanager tree, changes in both male and female plumage coloration contribute to the evolution of sexual dichromatism. The analyses of the complete dataset and the correlations that included the interclade contrasts indicate that male plumage evolution early in the history of the tanagers might have been the result of a change in female preference for particular aspects of male plumage, or a general change in life history. When different models of evolution were tested for the evolution of male and female coloration, the best-fit model for both sexes was the most complex model of light environment. Taken together, it seems that signaling is important in the evolution of male and female plumage coloration, perhaps to increase male conspicuousness and female crypsis.
