The activity budget hypothesis is 1 of 4 main hypotheses proposed to explain sexual segregation by large herbivores. Because of their smaller body size, females are predicted to have higher mass-specific energy requirements and lower digestive efficiency than males. As a result, females are expected to forage longer to satisfy their nutritional demands. Maintaining the cohesion of a mixed-sex group with differing activity budgets and asynchronous behavioral patterns is increasingly difficult, ultimately leading to spatial segregation of males and females. We tested this hypothesis using data (2002–2005) from 3 distinct populations of African elephants (Loxodonta africana), a species that exhibits marked sexual segregation. Group and individual behaviors were assessed at discrete points in time throughout the day, with a minimum of 10 min between consecutive records. Focal samples of individual male and female elephants also were recorded, with behavioral data logged every minute for 15 min. Data were grouped into 5 behavioral categories: drinking, resting, walking, feeding, and other. Neither activity rhythms nor feeding time varied significantly between the sexes and behavioral patterns were very similar. We propose that social and environmental factors influence behavioral rhythms to a greater extent than does body size, whereas increasing feeding time is only 1 method by which elephants can improve nutritional return. This is especially pertinent when considering their generalist foraging approach, substantial energy demands, and hindgut fermentation. We conclude that the activity budget hypothesis is unlikely to be the causal mechanism in the sexual segregation of African elephants, a finding that concurs with recent experimental and field research on a range of sexually dimorphic herbivores.