Spatial autocorrelation, phylogenetic constraints, and the causes of sexual dimorphism in primates

Abstract

Cheverud et al. (1985) apply the important and relatively new methodology of spatial autocorrelation to the quantification of phylogenetic constraints on adaptation and illustrate the use of these methods in an allometric study of sexual dimorphism in body size among extant nonhuman primates. Though of potentially broad applicability, the technique was completely overlooked in a recent review of methods to control for the effects of common descent in comparative studies (Bell, 1989). Their approach therefore deserves a wider recognition. However, their specific conclusion, that phytogeny is the primary determinant of patterns of sexual dimorphism among primates, has been uncritically accepted. We present four main methodological problems with their approach that should temper the interpretation of their analysis: biased phylogenetic relatedness scores, biased sample selection, size dependence in sex dimorphism measurement, and deficiencies in selection of a structural path model. We conclude that, even in terms of the analysis by Cheverud and co-workers (1985), phylogenetic inertia is not the primary reason for body size dimorphism.