Abstract
Male American lobster (Homarus americanus) from inshore southern New England were discriminated from offshore males on the basis of larger relative chela size. Lobsters from Buzzards Bay (inshore) had more conspicuous sexual dimorphism than lobsters from Hydrographer Canyon (offshore), and allometric growth of male chelae was more prominent than that of female abdomens. Principal components analysis of males from combined stocks represented variability in multivariate size and relative chela size, and component score distributions of each stock were discrete. Principal components of females from both stocks comprised variability in overall size and relative abdomen size, but principal component scores overlapped extensively. Multiple-group principal component 2 was a size-free index of relative chela size that classified 96% of males to the correct stock. Multiple-group principal component 2 of females did not successfully separate stocks. Discriminant analysis of size-adjusted morphometric data classified males to stock with 100% accuracy on the basis of relative chela size. Although discrimination of size-adjusted female data classified stocks with 94% accuracy, it was less stable and not associated with onset of maturity.