The shape of an animal’s hands provides insight not only to its locomotory habitus but can also reveal information about the genetic and developmental sources that underlie its variation. Detailed analyses of skeletal shape variation within a population can test hypotheses about the genetic, nongenetic, and epigenetic sources underlying that variation. Here, we report on the variation, patterns of correlation, and sexual dimorphism of human metacarpal size in order to better understand the evolutionary history of the hominid hand. Seven linear measurements were collected from unaffiliated adult Native Californians that lived between 3050 BP and 150 BP, correlations across digits were estimated and compared for the entire population, and for males and females. We also assessed sexual dimorphism in variance as well as for metacarpal length ratios.

Results indicate the thumb, or pollical metacarpal (MC1) measurements are only weakly correlated with those of the index through pinky fingers (palmar metacarpals, MC2-5), whereas all palmar metacarpals are more highly correlated with each other. The lower level of correlation between the pollical and palmar metacarpals accords with expectations from non-human developmental studies that indicate developmental modularity between these rays (and as a consequence, this results in developmental independence and the potential for selection to operate on the modules distinctly). Sexual dimorphism is observed in the absolute size of the metacarpals, and also in the degree of variation (males exhibit a greater range of variation) and level of correlation (females return lower correlations for the palmar metacarpal measurements). In contrast, metacarpal length ratios were not sexually dimorphic. The dimorphism in degree of variance and correlation raise new directions for research, while simultaneously bolstering the interpretation that human pollical and palmar metacarpals reflect two distinct developmental modules.