Publication TypeJournal Article
Year of Publication1994
AuthorsWalls, SC, Blaustein, AR
Type of ArticleJournal Article

We examined the effects of variation in larval density, genetic relatedness and their interaction in the marbled salamander, Ambystoma opacum. In a factorial experiment, we reared larvae in artificial ponds at low and high densities and in groups comprised of either a single sibship (all siblings) or an equal mixture of six different sibships. After 150 d (before the onset of metamorphosis), we measured the proportion of larvae surviving, body size (snout-vent length and mass), and the size distribution of larvae in each pond. High initial densities significantly reduced body mass, but had no significant effect on larval snout-vent length, the size distributions of individuals, or survival. Neither variation in genetic relatedness (single vs mixed sibship groups), nor its interaction with initial density, influenced larval performance. Because larval survival was independent of initial density, we also assessed whether body size variables were primarily a function of the final number of survivors in each pond. Simple linear regression analyses revealed that larval snout-vent length and mass were negatively related to final density, whereas the skew in the size distributions of individuals increased with increasing final density. Separate regression analyses for groups of siblings vs groups of mixed relatedness revealed no significant differences in the regression slopes and intercepts for these two groups, thus indicating that both estimated the same regression population. Therefore, kinship did not appear to influence density-dependent larval performance in the environment of our artificial ponds. Our results contrast with evidence of kinship effects on behavior under laboratory conditions for this and other species of Ambystoma and illustrate the context-dependency With which kin recognition may operate under variable experimental conditions.

URL<Go to ISI>://WOS:A1994QA75600013