Experience-dependent courtship modification in the housefly, Musca domestica

The genetics of mating behavior may be complicated by factors such as an association with fitness, social interactions, and genotype-by-environment interactions. In order to evaluate the potential consequences of these factors, I analyzed courtship behavior in the housefly, Musca domestica L. by specifically examining (1) mating success, (2) experience-dependent courtship modification (learning), and (3) heritability. Males (N = 120) were paired with six virgin females over the course of three days and videotaped for a maximum of 30 minutes, or until copulation occurred. Courtships were analyzed for six behaviors within and among females. Successful courtships, in comparison to unsuccessful attempts, were characterized by a shorter Duration of Courtship, less intense Lunge and Wingout, and more intense Buzz, Hold, and Lift. Among courtship attempts within females, males decreased Buzz and increased Lunge. The females also adjusted their behavior by modifying Wingout, which demonstrates a male-female interaction component of mating success. Among courtship attempts across serially mated females, males decreased the Duration of Courtship and Lunge. Analyses were also completed using both within and across female data, and there was a general trend for a decrease in the behavior traits, except for Buzz, which did not change. Also, males decreased the number of courtship attempts with successively mated females, indicating the ability to learn through improved courtship technique. Weibull functions showed that the males had significant learning curve functions for individual behavior traits, although the tests for the repeatability and heritability of such functions were not significant. Finally, repeatabilities and heritabilities (estimated by a broad-sense family effect) were calculated for the courtship traits. Pooled analyses on successful and unsuccessful courtship attempts revealed significant repeatabilities (0.05-0.67) and heritabilities (0.07-0.50) for almost all of the courtship traits, although analyses on just the successful courtships failed to replicate the significant effects, except for the male display, Hold (0.29). These results demonstrate that male courtship behavior is multifaceted in that it consists of not only a learning component, but a genetic component, too. Although previous housefly experiments have shown indirect support, this is the first direct evidence for male learning within courtship behavior