Ecological causes for the evolution of sexual dimorphism: a review of the evidence.

Can sexual dimorphism evolve because of ecological differences between the sexes? Although several examples of this phenomenon are well known from studies on birds, the idea has often been dismissed as lacking general applicability. This dismissal does not stem from contradictory data so much as from the difficulties inherent in testing the hypothesis, and its apparent lack of parsimony, in comparison to the alternative explanation of sexual selection. The only unequivocal evidence for the evolution of sexual dimorphism through intersexual niche partitioning would be disproportionate dimorphism in trophic structures (e.g., mouthparts). This criterion offers a minimum estimate of the importance of ecological causes for dimorphism, because it may fail to identify most cases. A review of published literature reveals examples of sexually dimorphic trophic structures in most animal phyla. Many of these examples seem to be attributable to sexual selection, but others reflect adaptations for niche divergence between the sexes. For example, dwarf non-feeding males without functional mouthparts have evolved independently in many taxa. In other cases, males and females differ in trophic structures apparently because of differences in diets. Such divergence may often reflect specific nutritional requirements for reproduction in females, or extreme (sexually selected?) differences between males and females in habitats or body sizes. Ecological competition between the sexes may be responsible for intersexual niche divergence in some cases, but the independent evolution of foraging specializations by each sex may be of more general importance. If ecological causation for dimorphism can be demonstrated in so many cases, despite the inadequacies of the available criteria, the degree of sexual size dimorphism in many other animal species may well also have been influenced by ecological factors. Hence, it may be premature to dismiss this hypothesis, despite the difficulty of testing it.