ANIMAL BEHAVIOUR, 1998, 56, 275–287 Article No. ar980784 ARTICLES Stabilizing and directional female choice for male calls in the European green toad SERGIO CASTELLANO & CRISTINA GIACOM A Dipartimento di Biologia Animale, Universita ` di Torino (Received 21 April 1997; initial acceptance 3 June 1997; final acceptance 27 September 1997; MS. number: 5513) ABSTRACT We analysed variation in three temporal (pulse rate, call and intercall duration) and one spectral (fundamental frequency) acoustic properties of the green toad’s, Bufo viridis, advertisement call, at different levels, from the single bout of an individual to between populations. Independent of the level, pulse rate and fundamental frequency were less variable than intercall and call durations. The former were classified as static, the latter as dynamic properties of the signal. We analysed the effects of body size and temperature, assumed to represent, respectively, morphological and physiological constraints on the signals. Static properties were under strong constraints, either morphologically (fundamental frequency) or physiologically (pulse rate), whereas no significant effects of size and temperature were observed on dynamic properties. We used two-choice discrimination experiments to investigate females’ preferences for call properties. Females preferred extreme to mean values of the dynamic properties, even when these fell above the typical range of variation of the population. In contrast, females had weak preferences for mean values, and for lower than average compared with higher than average values, of static properties. We discuss the hypothesis that static and dynamic call properties may convey different information to females: static properties may be important for species recognition, whereas dynamic properties may be important for mate selection.  1998 The Association for the Study of Animal Behaviour In many species of frogs and toads acoustic communi- cation is the most important means of mate recognition, females responding phonotactically to the vocalizations of males (Gerhardt 1988 , 1994 ). Two functions of male advertisement calls are usually recognized: to facilitate species recognition and to provide females with criteria for choosing their partner among many males. In the past, species recognition and sexual selection have been viewed as antithetical theories. In particular, since species recognition is supposed to generate strong stabilizing selection, whereas sexual selection is supposed to act directionally, some authors have believed that species recognition and sexual selection are mutually exclusive (Templeton 1979 ; Paterson 1985 ). More recent work has overcome this dichotomy and proposed a unitary theor- etical framework. Ryan & Rand (1993) suggested that species recognition and sexual selection can result from an interaction between variation in signals and the responses of females to this variation. They distinguished between the concepts of mate recognition and mate preference: their experiments showed that Physalaemus pustulosus females are sometimes attracted by hetero- specific calls when these are the only signals perceived, but when given a choice between heterospecific and conspecific calls they always choose (prefer) the latter (but see Ryan & Wagner 1987 and Basolo 1990 for heterospecific preferences). In these cases, therefore, both heterospecific and conspecific calls are recognized, but the latter are preferred. The probability that an acoustic stimulus will elicit a female phonotactic response is a function of the properties of the signal. Species recog- nition and sexual selection are described by different probability distribution functions (the so-called ‘female preference function’: Ryan & Rand, 1993 ), but they are not the result of fundamentally different biological processes. Female preference functions that result in Correspondence: S. Castellano, Dipartimento di Biologia Animale, Universita ` di Torino, Via Accademia Albertina, 17, 10123-Torino, Italy. 0003–3472/98/080275+13 $30.00/0  1998 The Association for the Study of Animal Behaviour 275