The chin, or mentum osseum, is one of the most distinctive anatomical traits of modern humans. A variety of hypotheses for the adaptive value of the chin have been proposed, ranging from mechanical stress resistance to sexual selection via mate choice. While the sexual selection hypothesis predicts dimorphism in chin shape, most biomechanical hypotheses preclude it. Therefore determining the presence or absence of significant sexual dimorphism in chin shape provides a useful method for differentiating between various adaptive hypotheses; however, this has yet to be done due to a lack of quantitative data on chin shape. The goals of this study are therefore: (1) to introduce a new method for quantifying chin shape and (2) to determine the presence or absence of sexual dimorphism in chin shape in a diverse sample of modern humans. Samples were drawn from recent human skeletal collections representing nine geographic regions. Outlines of mentum osseum contours were quantified using elliptical Fourier function analysis (EFFA). Fourier coefficients were analyzed using principal components analysis (PCA). Sexual dimorphism in chin shape was assessed using PC loadings in the pooled geographic sample, and statistically significant differences were found. These findings provide the first quantitative, morphologically based evidence in support of adaptive hypotheses that predict dimorphism in chin shape, including the sexual selection hypothesis. Am J Phys Anthropol 143:417-425, 2010. © 2010 Wiley-Liss, Inc.