Mating advantage of rare males in models of sexual selection

Models of sexual selection in polygynous species of animals have been derived on the assumption that some females have preferences to mate with males with particular genotypes. The mating advantage gained by the males is always frequency-dependent because the preferred males take part in the same number of preferential matings when they are rare as when they are common; individually therefore, they mate more often when they are rare. Frequency-dependent sexual selection has been demonstrated in many experiments with Drosophila: rare males take part in a higher proportion of matings than their frequency as available mates. Ehrman and Spiess explained this phenomenon by frequency-dependence either in female preference or in male courtship. This explanation, which is difficult to interpret in behavioural terms, may not be necessary, however, because constant female preferences would entail frequency-dependent selection among the males. I show here that a simple model of constant preferences for particular phenotypes or genotypes is sufficient to explain a large body of data on frequency-dependent sexual selection in Drosophila.