Courting females: ecological constraints affect sex roles in a natural population of the blenniid fishSalaria pavo

Abstract

In most animal species females are the sex with the limiting reproductive rate, which promotes the predominance of male intra-sexual competition, active courtship and more developed epigamic characters in males (Trivers 1985). In some species, however, sex-role reversal occurs, and the females actively court and compete for mates (insects: Gwynne 1981; Svensson & Petersson 1987; stomatopods: Hatziolos & Caldwell 1983; poison-arrow frog: Wells 1978; birds: Petrie 1983; Reynolds et al. 1986; Colwell & Oring 1988; pipefish: Berglund et al. 1989; Berglund & Rosenqvist 1990, 1993; Rosenqvist 1990). In these cases the operational sex ratio is biased towards females so that the reproductive rate is higher in females, often because of special features of male reproductive biology (e.g. pipefish; Berglund et al. 1989). In this paper we present evidence for ecological modulation of sex roles in a natural population of the blenniid fish Salaria pavo (Risso). While in other known populations of this species parental males defend nesting territories where they actively court females (Fishelson 1963; Patzner et al. 1986), in the population studied, where nest sites are very scarce, courtship is almost entirely initiated by females, and both male and female intra-sexual competition occurs. Males of S. pavo establish nests in natural cavities where they guard the eggs. Several females may visit a single nest and each female may spawn successively with different males during the breeding season (Patzner et al. 1986). The courtship in this species is usually initiated by the male (lateral jerking, zig-zag swimming and figure-8 swimming) and receptive females respond with a specific display (nuptial coloration, rapid respiratory movements, pectoral fin fanning and exposure of the ventral region to the male; Patzner et al. 1986). The population we studied occurs in a lagoon (Ria Formosa, southern Portugal), where the only available cavities are in bricks used by clamculturists to delimit their fields. Only those brick holes that have one of the openings obstructed serve as potential nest sites, and some bricks have several such holes. In such bricks guarding males form dense nest aggregations and do not defend a territory around the nest site. There is an excess of mature males that cannot establish nests and nesting males are significantly larger than nonnesting males. A detailed description of this population and its ecology is provided in Almada et al. (1994). Behavioural observations on 97 nests were conducted by snorkelling in the breeding seasons of 1989 and 1994 (total observation time=28 h, 54 min; each observation lasted from 15 to 90 min with an average of 25 min). In each observation we recorded the following variables: number of nests in the brick; number of visits and sex of the visitor; number of agonistic interactions and the identity (parental, floater or female) of the participants, and number of courtship episodes and the sex of the initiator. There was a high level of competition for nest sites among the males. Intruding males were often observed to threaten and sometimes bite the protruding head of the guarding males, and most agonistic interactions between guarding and intruding males were initiated by the latter (chisquared goodness-of-fit test assuming equal frequencies: ÷=20·6, df=1, P<0·001). Removal experiments confirmed the intense male competition for nests. In 15 nests in which the parental male was removed, 10 had been reoccupied by another male by the following day. This reoccupation rate is much higher than that found in a natural population of another intertidal blenny