Evolution of female remating behaviour following experimental removal of sexual selection

The relatively small number of ova produced by a female can be fertilized by a single ejaculate in most species. Why females of many species mate with multiple males is therefore enigmatic, especially given that costs associated with remating have been well documented. Recently, it has been argued that females may remate at a maladaptive rate as an outcome of sexually antagonistic coevolution: the evolutionary tug-of-war between manipulation by one sex and resistance to being manipulated by the other sex. We tested this hypothesis experimentally for the evolution of the female remating interval in a naturally promiscuous species, Drosophila melanogaster. In two replicate populations, sexual selection was removed through enforced monogamous mating with random mate assignment, or retained in polyandrous controls. Monogamy constrains the reproductive success of mates to be identical, thereby converting prior conflicts between mates into opportunities for mutualism. Under these experimental conditions, the sexually antagonistic coevolution hypothesis generates explicit predictions regarding the direction of evolutionary change in female remating behaviour. These predictions are contingent upon the mechanism of male manipulation, which may be mediated biochemically by seminal fluids or behaviourally by courtship. Levels of divergence in female remating interval across lines, and in male ejaculatory and courtship effects on female remating, were quantified after 84 generations of selection. Data refute the hypothesis that the evolutionary change in female remating behaviour was due to sexually antagonistic coevolution of courtship signal and receiver traits. The data were, however, consistent with a hypothesis of sexual conflict mediated through ejaculate manipulation. Monogamy-line males evolved ejaculates that were less effective in inducing female non-receptivity and monogamy-line females evolved to remate less frequently, symptomatic of lowered resistance to ejaculate manipulation. The consistency of the results with alternative hypotheses to explain female promiscuity are discussed.