Female mate choice under predation risk in the guppy

Sexual selection, and thus the evolution of both female mating preferences and preferred male traits, may be constrained by costs associated with mate choice. Recent theoretical models predict that female preference should decrease with increasing costs of mate choice such as predation risk. This general hypothesis was tested experimentally by presenting individual female guppies, Poecilia reticulata, originating from two Trinidadian populations with different predation regimes, with a choice of paired males (differing mainly in colour pattern) as potential mates in both the absence and presence of a cichlid fish predator that naturally co-occurs in one of these populations. Moreover, any reduction in female preference under risk was predicted to be positively correlated with the level of predation intensity occurring in their respective populations. In an experiment in which only the female could see the nearby predator, females from the high-predation population (Quare River) significantly reduced both their overall level of sexual activity and their preference for a particular male (most commonly the more colourful male) in the presence of the predator. In contrast, the sexual activity level and mating preference of females from the low-predation population (Paria River) were unaffected by the apparent threat of predation. Observed changes in female sexual activity level and mating preference could not have been mediated by any change in male behaviour when the predator was present, because the males could neither see or smell the predator nearby. In a corollary experiment in which both males and females from the risk-sensitive Quare River population could freely interact with one another and see the predator, females significantly reduced their preference for the more brightly coloured male in the presence of the predator, irrespective of predator-mediated changes in male courtship behaviour. These results are best explained by a reduction in female preference under increased apparent risk of predation, and are consistent with the general hypothesis that female mate choice in the guppy is potentially costly in terms of an increased risk of mortality due to predation and that females from different populations are differentially sensitive to this risk when assessing and choosing potential mates. Therefore, predators may exert direct selection on female mate choice in certain natural populations of the guppy. (C) 1996 The Association for the Study of Animal Behaviour