Evolutionary Ecology 1996, 10, 97-104 Sexual selection and the allometry of earwig forceps L.W. SIMMONS and J.L. TOMKINS Department of Environmental and Evolutionary Biology, University of Liverpool, POBox 147, Liverpool, L69 3BX, UK Summary Positive intraspecific allometry, the tendency for large individuals to have relatively larger morphological traits, is thought to be more likely for secondary sexual traits than naturally selected traits. This is because secondary sexual traits are often used to signal individual quality and positive allometry should arise where the costs and/or benefits of signalling are size dependent. Here we examine the allometric relationships between forceps length, a sexually selected trait and elytra length, a naturally selected trait, in 42 species of earwig. Both forceps and elytra showed positive allometry. However, the degree of allometry was greater for forceps as predicted, ff ailometry arises due to sexual selection we would predict a greater degree of allometry in species with more exaggerated secondary sexual traits. Across species, the degree of forcep allometry did increase with forcep exaggeration. The relevance of positive allometry to reliable signalling is discussed. Keywords: Dermaptera; secondary sexual traits; positive allometry; display Introduction In his study of Megaloceros antlers, Gould (1973a) interpreted positive intraspecific allometry, the tendency for larger individuals to possess relatively larger morphological traits, as an evolutionary consequence of status signalhng by males, both to competitors and potential mates. Petrie (1988, 1992) proposed that in general positive allometry is more likely to occur in secondary sexual traits than in other naturally selected traits. Because the outcome of competitive interactions is usually determined by body size (Davies and Halliday, 1979; Sigurj6nsd6ttir and Parker, 1981; Petrie, 1984; Simmons, 1986) it should pay large individuals to produce large secondary sexual traits that signal competitive ability as this would reduce the number and/or intensity of agonistic interactions in which they become involved. The same argument does not hold for small individuals because they have little to gain from advertising low competitive ability and, further, any costs associated with developing the trait are likely to outweigh the benefits of signalling competitive ability. In general it has been argued that sexually selected traits may act as handicaps in the sense that larger individuals may be better able to withstand the costs associated with producing them (Zahavi, 1975; Andersson, 1986). Thus, the size of secondary sexual traits may represent an honest signal of an individual's competitive ability (Parker, 1974; Alatalo et al., 1988; Petrie, 1988). Green (1992) argued that secondary sexual traits that function in mate choice should similarly show positive allometry, for example, if smaller individuals are less able to find mates or are in lower phenotypic condition, they should invest relatively less in secondary sexual displays required for mate attraction. Positive allometric relationships have been found in the secondary sexual traits of a variety of taxa (Huxley, 1931; Gould, 1973h; Brown and Bartalon, 1986; Alatalo et al., 1988; Petrie, 1988). Nevertheless, positive allometry can be predicted for a variety of reasons and need not be specific to secondary sexual traits. Although Green (1992) found a positive allometric relationship for tail height, a sexually selected trait in male smooth newts, he also found positive allometry for male 0269-7653 © 1996 Chapman & Hall