We tested the validity of Bergmann's rule and Rosenzweig's hypothesis through an analysis of the geographical variation of the skull size of Otaria flavescens along the entire distribution range of the species (except Brazil). We quantified the sizes of 606 adult South American sea lion skulls measured in seven localities of Peru, Chile, Uruguay, Argentina, and the Falkland/Malvinas Islands. Geographical and environmental variables included latitude, longitude, and monthly minimum, maximum, and mean air and ocean temperatures. We also included information on fish landings as a proxy for productivity. Males showed a positive relationship between condylobasal length (CBL) and latitude, and between CBL and the six temperature variables. By contrast, females showed a negative relationship between CBL and the same variables. Finally, female skull size showed a significant and positive correlation with fish landings, while males did not show any relationship with this variable. The body size of males conformed to Bergmann's rule, with larger individuals found in southern localities of South America. Females followed the converse of Bergmann's rule at the intraspecific level, but showed a positive relationship with the proxy for productivity, thus supporting Rosenzweig's hypothesis. Differences in the factors that drive body size in females and males may be explained by their different life-history strategies. Our analyses demonstrate that latitude and temperature are not the only factors that explain spatial variation in body size: others such as food availability are also important for explaining the ecogeographical patterns found in O. flavescens. © 2012 Springer-Verlag.