BackgroundThe evolution of exaggerated vocal signals in anuran species is an important topic. Males and females have both evolved the ability to discriminate communication sounds. However, the nature of sexual dimorphism in cognition and sensory discrimination and in the evolution and limitation of sexual signal exaggeration remain relatively unexplored.MethodsIn the present study, we used male calls of varied complexity in the serrate-legged small treefrog, Kurixalus odontotarsus, as probes to investigate how both sexes respond to variations in call complexity and how sex differences in signal discrimination play a role in the evolution of sexual signal exaggeration. The compound calls of male K. odontotarsus consist of a series of one or more harmonic notes (A notes) which may be followed by one or more short broadband notes (B notes).ResultsMale playback experiments and female phonotaxis tests showed that increasing the number of A notes in stimulus calls elicits increased numbers of response calls by males and increases the attractiveness of the stimulus calls to females. The addition of B notes, however, reduces male calling responses. Moreover, call stimuli which contain only B notes suppress spontaneous male calling responses. Phonotaxis experiments show that females prefer calls with greater numbers of A notes and calls containing both A notes and B notes, but do not prefer calls with only B notes.DiscussionMale-male competition and female choice appear to have played different roles in the evolution and limitation of signal complexity in K. odontotarsus. These results provide new insights into how exaggerated compound signals evolve and how signal complexity may be limited in anurans.

Most species are believed to evolve larger body sizes over evolutionary time. Previous studies have suggested that sexual selection, through male-male competition and female choice, favors larger males. However, there is little evidence of selection against large size. The female serrate-legged small treefrogs (Philautus odontotarsus) must carry passive males from leks to breeding grounds over relatively long distances after amplexus to find a suitable place to lay eggs. The costs of large male size may therefore decrease mating success due to reduced agility and/or higher energy requirements. Thus, we hypothesized that selection would not favor larger males in P. odontotarsus. Females can assess male body size on the basis of the dominant frequency of male calls in frogs. To assess female P. odontotarsus preferences for a potential mate’s body size, male calls of high, average and low dominant frequency were played back to the females in phonotaxis experiments. Results showed that most females prefer the advertisement call with average dominant frequency. In addition, we compared the body mass distribution of amplectant males with that of single males in nature. The body masses of amplectant males are more narrowly distributed in the intermediate range than that of single males. The phonotaxis results and the data of actual female preferences in the field show that females strongly prefer potential mates of mean body sizes, consistent with the view that, in this species at least, larger males are not always perceived as better by females. In the present study, P. odontotarsus provides an example of an amphibian species in which large size does not have an advantage in mating success for males. Instead, our results provide evidences that stabilizing selection favors the optimal intermediate size of males.