Sexual signal loss in field crickets maintained despite strong sexual selection favoring singing males.

Evolutionary biologists commonly seek explanations for how selection drives the emergence of novel traits. Although trait loss is also predicted to occur frequently, few contemporary examples exist. In Hawaii, the Pacific field cricket (Teleogryllus oceanicus) is undergoing adaptive sexual signal loss due to natural selection imposed by eavesdropping parasitoids. Mutant male crickets ("flatwings") cannot sing. We measured the intensity of sexual selection on wing phenotype in a wild population. First, we surveyed the relative abundance of flatwings and "normal-wings" (nonmutants) on Oahu. Then, we bred wild-mated females' offspring to determine both female genotype with respect to the flatwing mutation and the proportion of flatwing males that sired their offspring. We found evidence of strong sexual selection favoring the production of song: females were predominantly homozygous normal-wing, their offspring were sired disproportionately by singing males, and at the population level, flatwing males became less common following a single sexual selection event. We report a selection coefficient describing the total (pre- and postcopulatory) sexual selection favoring normal-wing males in nature. Given the maintenance of the flatwing phenotype in Hawaii in recent years, this substantial sexual selection additionally suggests an approximate strength of opposing natural selection that favors silent males.

Alternative Reproductive Tactics Arising from a Continuous Behavioral Trait: Callers versus Satellites in Field Crickets.

Alternative reproductive tactics may arise when natural enemies use sexual signals to locate the signaler. In field crickets, elevated costs to male calling due to acoustically orienting parasitoid flies create opportunity for an alternative tactic, satellite behavior, where noncalling males intercept females attracted to callers. Although the caller-satellite system in crickets that risk detection by parasitoids resembles distinct behavioral phenotypes, a male's propensity to behave as caller or satellite can be a continuously variable trait over several temporal scales, and an individual may pursue alternate tactics at different times. We modeled a caller-satellite-parasitoid system as a spatially explicit interaction among male and female crickets using individual-based simulation. Males varied in their propensity to call versus behave as a satellite from one night to the next. We varied mortality, density, sex ratio, and female mating behavior, and recorded lifetime number of mates as a function of a male's probability of calling (vs. acting as a satellite) along a gradient in parasitism risk. Frequently, the optimal behavior switched abruptly from being pure caller (call every night) to pure satellite (never call) as parasitism rate increased. However, mixed strategies prevailed even with high parasitism risk under conditions of higher background mortality rate, decreasing density, increasing female-biased sex ratio, and increasing female choosiness. In natural populations, high parasitoid pressure alone would be unlikely to yield fixation of pure satellite behavior.