SEXUAL SELECTION IN THE GULF-COAST TOAD, BUFO VALLICEPS - FEMALE CHOICE-BASED ON VARIABLE CHARACTERS

Sexual selection on male size and calling behaviour was studied for 2 years in one population and for 3 years in a second population of the Gulf Coast toad. A large-male mating advantage was detected only at on e site during 1 year. Measures of calling behaviour were available for 3 years at one of the sites: on a nightly basis in 2 years, males wit h higher call rates were significantly more likely to mate than males with lower call rates. However, there was no relationship between mati ng success and either call duration or calling effort on a nightly bia s in any year. In discrimination experiments, females preferred higher call rates and calls of longer duration, but not calls of lower frequ ency. Biased male mating success with respect to call rate therefore a ppears to have been partially a consequence of female mate choice. Cal l duration is positively correlated with male size, while dominant fre quency is negatively correlated with male size in this species. Becaus e females did not discriminate between high- and low-frequency calls, and because there was no bias in mating success with respect to call d uration, it appears unlikely that female choice resulted in the large- male mating advantage observed in 1 year. Despite a nightly bias in ma ting success with respect to call rate that appears to have resulted f rom female mate choice, there was no detectable sexual selection an av erage call rate in any year. Instead, the number of nights of chorusin g activity was the only detectable direct target of selection in all 3 years at one of the sites. The repeatability of call rate within male s was low, ranging from 0.08 to 0.33. Simulations suggest that within- male variation in a trait can reduce the strength of sexual selection on the trait. Lack of selection on call rate, despite female choice ba sed on call rate, may thus have been partially a consequence of within -male variation in this trait.