Mating among birds is usually a matter of mutual agreement: both male and female are being selective in their choice of mates (although their criteria are likely to differ). Selectivity in mate choice is central to the Darwinian process of sexual selection, determining which individuals will mate and produce offspring. The chosen individuals are more successful in mating and reproduction. Thus, their genes are passed on to the next generation while the genes of un-chosen individuals are not. As the process continues, the genes that determine both the winning traits and the choosiness become more frequent in the population.
The outcomes of all that choosing vary enormously, depending on the ecology of the species, the previous evolutionary history, and the occurrence of genetic variation upon which selection can happen. If there is no plumage variation, there is nothing to choose! Avian feathers serve various functions, one of which is visual display during courtship. Genetic variation in plumage among potential mates provided birds with choices of color and pattern as they decide with whom to mate. Those choices, together with variations in other traits, shape the appearance and behavior of the lineage.
Imagine a species (Species A) that makes open-cup nests (as do all related species…that’s an evolutionary history factor) in shrubs (instead of under logs or roots; that’s an ecology factor). Suppose that in this species it is (for whatever historical reason) the female that incubates the eggs and cares for the nestlings. In an open-cup nests (compared to a cavity nest, for instance), the incubating female would be exposed to searching predators, and her frequent feeding visits to a particular location would also be noted by lurking predators. So inconspicuous, perhaps camouflaged, plumage would lead to better survival of the female and the eggs and chicks. And a male that preferred an inconspicuous female would have better reproductive success than one that somehow preferred a female that flashes a noticeable red crest or a long, bright blue tail.
If there is some variation among the males in the colorfulness of their plumage, those drab females might prefer males that are a bit colorful, perhaps with a yellow head or vivid magenta wings, rather than males that are more like the females. If those plumage patterns are heritable, the sons and daughters of such adults would bear the same traits, and eventually the whole population would have drab females and colorful males.
However, if (in species B) the males also do some of the incubation and parental care, they and their eggs and nestlings might suffer more predation. If they were colorful, then they and any female that somehow preferred a gaudy male would probably have lowered reproductive success. And so their genes would become less frequent in the population, and males and females would look similar in color.
Now go back to species A. Imagine that the population is spread over quite a large geographic area, such that the birds in one area just don’t get to another area, or vice versa. Now it is possible for the birds in that area to become different from the rest of the population. Suppose that some males don’t have simple yellow heads, they have additional blue crests on top. Then females there might find that they like males with blue crests (instead of plain yellow heads), for example. Then these more fancy males would come to predominate in that area. That might happen in several different areas, with different outcomes in each. If some of these fancy males just happened to wander into a different area, they might not be preferred by the females there. So thus, species A has begun to diversify into several new species, each with different male ornaments (and female preferences).
The classic example of diversification driven principally by mate choice is in the neotropical manakins. There are over fifty species of manakins. In general, the females are greenish and plain, while the males sport a spectacular array of plumage patterns and colors, often with behaviors that show off those features. By being very choosy, females maintain the dramatic differences among the males, and interbreeding is rare to non-existent.
Another example might be the wood warblers of North America, in which the males of different species are generally somewhat more colorful and distinctive than the females. Again, interbreeding between species is not common, but in a few cases, hybridization occurs between two species (e.g., Townsend’s and hermit warblers). Here in Juneau, a few years ago, observers noted that a warbler nest was tended by a mixed set of parents belonging to the same species (yellow-rumped warbler) but of two different varieties. One seemed to conform to the plumage patterns of Audubon’s warbler, while the other one was either a typical myrtle warbler, or the result of a previous mixed-mating of the two forms. In either case, more hybrids were being produced by this pair.
Evolution by mate choice is common and widespread among birds. However, two other kinds of mating behavior tend to obscure the typical patterns. The first is that, even among ostensibly monogamous pairs, both sexes may go gallivanting, and do some of their copulations outside of the pair bond, and broods of mixed parentage occur. The choices for intra-pair copulations and extra-pair copulations may or may not be the same.
A second kind of mating behavior totally subverts the normal patterns of mate choice. In many ducks and some geese, there are forced copulations, in which males attack and try to copulate with females, which struggle and resist, but commonly suffer injury (sometimes lethal). There is clearly no female preference involved. The extraordinary complexity of the female reproductive tract in these species probably evolved as a way of reducing the fertilization success of the forced copulations; nevertheless, some small percentage of the embryos can be fathered by these violent males. Injury to the violated females is likely to reduce their nesting success, but I have not found data on that. The origin and continuation of this behavior of males is not entirely clear.