Rapid evolutionary change over a few generations has been documented in natural populations. Such changes are observed as organisms invade new environments, and they are often triggered by changed interspecific interactions, such as differences in predation regimes. However, in spite of increased recognition of antagonistic male-female mating interactions, there is very limited evidence that such intraspecific interactions could cause rapid evolutionary dynamics in nature. This is because ecological and longitudinal data from natural populations have been lacking. Here we show that in a color-polymorphic damselfly species, male-female mating interactions lead to rapid evolutionary change in morph frequencies between generations. Field data and computer simulations indicate that these changes are driven by sexual conflict, in which morph fecundities are negatively affected by frequency-and density-dependent male mating harassment. These frequency-dependent processes prevent population divergence by maintaining a female polymorphism in most populations. Although these results contrast with the traditional view of how sexual conflict enhances the rate of population divergence, they are consistent with a recent theoretical model of how females may form discrete genetic clusters in response to male mating harassment.

Evolutionary conflict between the sexes is predicted to lead to sexual arms races in which male adaptations for acquiring mates ("offense" traits) are met by female counteradaptations--for example, to reduce mating rate ("defense" traits). Such coevolutionary chases may be perpetual. However, we show here that the coevolutionary process may also lead to a stable state in which multiple offense-defense trait pairs are maintained. This type of polymorphism below the species level is a result of sexual conflict in combination with nonrandom mating. Our results show that if nonrandom mating occurs with respect to male and female conflict traits, genetic correlations will act to stabilize the trait frequencies so that all morphs are maintained. We discuss the results in special relation to the evolution of female polymorphism in diving beetles and argue that the process we describe may be a general force that maintains polymorphism in other taxa as well.