Variation in safety factors of claws within and among six species of Cancer crabs (Decapoda : Brachyura)

To better understand how safety factors of biological structures evolve, we examined the frequency of claw failure, and the intra- and interspecific p atterns of variation in maximum biting force and breaking strength in the c laws of six species of Cancer (Linnaeus) crabs that live in sympatry along the coast of the northeastern Pacific: C. antennarius, C. branneri, C. grac ilis, C. magister, C. oregonensis and C. productus. Although the breakage f requencies in natural populations were similar among species (approximate t o 6%), they were higher than predicted based on failure probabilities calcu lated from laboratory measurements of biting force and breaking strength fo r healthy pristine claws. The incidence of claw damage was correlated with the degree of wear, suggesting that claws later in the intermolt interval w ere more likely to fail. Within species, safety factors increased from 3.1 to 4.6 with increasing instar number due primarily to a decline in muscle s tress (force per unit area of apodeme). Surprisingly, the lower maximum mus cle stress generated by later instars appeared to be due to behavioral rest raint, since it was not accompanied by relatively lower muscle mass. In add ition, among individuals of the same claw size, lower breaking forces were correlated with lower maximum hiring force, and both were correlated with l ighter cuticle and closer muscle mass, suggesting a coupling that maintains a more stable safety factor over the moult cycle. In some species, size-ad justed maximum biting forces were higher for males than females, but this p aralleled differences in breaking strength, so safety factors did not diffe r between the sexes. Among the six Cancer species, one exhibited an unusual ly high safety factor (C. oregonensis, 7.4) and another an unusually low on e (C. magister, 2.6). The remaining four species were similar to each other and exhibited an intermediate safety factor (3.6). From a phylogenetic per spective, the species with more extreme safety factors appeared to be deriv ed from a common ancestor with an intermediate safety factor. From an ecolo gical perspective, species more closely associated with rocky sui,strata, a nd presumably a higher incidence of hard-shelled prey, exhibited higher saf ety factors. But safety factors were also correlated with relative claw siz e, and sexual dimorphism in claw size. Although we cannot say whether habit at, diet or sexual selection are primarily responsible for the differences in safety factors observed among species, the cost of producing a relativel y larger claw seems an unlikely explanation because safety factors did not differ between males and females in any of the sexually dimorphic species. (C) 2000 The Linnean Society of London.