Previously undocumented regional variability in crab-eating macaque skull sexual dimorphism and its implications for...

In a large sample of adult crab-eating macaques, we quantified sexual dimorphism in size, shape, and covariance across the whole skull and among anatomical regions of the cranium and mandible. All regions showed significant mean sex differences, but the magnitude of size and shape dimorphism varied substantially. Shape dimorphism displayed particularly pronounced extremes of variation in specific regions. Males were consistently more variable than females. However, sex differences in size variance were significant, whereas differences in shape variance were typically nonsignificant. The chondrocranium deviated from this pattern, exhibiting an unusually large and significant difference in shape variance. Almost all regions differed significantly in shape covariance structure between sexes. To test whether these dimorphisms could be biased by the subsetting of landmarks, we compared each anatomical region with random subsets of equal dimensionality. Two regions, the vertical ramus of the mandible and the chondrocranium, emerged as unusually weakly dimorphic in both size and shape, unlike most other regions whose dimorphism was marked but within expectations from random subsets. The horizontal ramus showed the strongest shape dimorphism, but also in this case the result was not greater than found among random subsets. These analyses not only report previously undocumented regional variability in sex differences but also emphasize how between sex heteroscedasticity can be substantial. Such differences are biologically informative but suggest limitations to evolutionary comparisons by challenging common morphometric practices, such as statistical "sex-corrections" and pooling of sexes to increase sample size, as they may lead to misleading results.