Sexual dimorphism in the walrus mandible: comparative description and geometric morphometrics

Abstract

The modern walrus Odobenus rosmarus is characterized by marked sexual dimorphism, related to its polygynous behavior and the aggressive competition between males during the breeding season. Previous studies treated skeletal sexual dimorphism in walruses either qualitatively or with basic quantitative measurements. The present study combines a detailed qualitative comparison of male and female walrus mandibles with quantitative two-dimensional geometric morphometrics analysis (principal component analysis, Procrustes ANOVA and a linear discriminant analysis). In addition to identifying previously recognized sexually dimorphic features (e.g., convexity of the anterior margin of the mandible in adult males), our study finds new morphological differences between males and females, such as a relative dorsal expansion of the anterior part of the mandible and an accentuated concavity between the dorsal margin and the coronoid process in adult males. Both our qualitative comparisons and quantitative analyses demonstrate that sexual dimorphism as expressed in the mandible of extant walruses is statistically significant and that (variation in) mandibular morphology can be used as tool to attribute sex with a good degree of accuracy to isolated mandibles or skeletons lacking the cranium. Sexual dimorphism in walruses is directly related to their sexual behavior, characterized as aggressive in males and linked to a polygynous reproduction system. Indeed, the difference in size of the tusks between males and females but also the use of these during intraspecific fights, can reasonably account for this great mandibular morphological disparity between adult males and females, but also among different ontogenetic stages. Finally, the results obtained in the present study may serve as a starting point for assessing sexual dimorphism more in-depth and studying inter- and intraspecific variation in the mandibles of fossil walruses by identifying quantified size and shape mandibular features.