THE EVOLUTION OF REVERSED SEXUAL DIMORPHISM IN SIZE IN MONOGAMOUS SPECIES OF BIRDS

Abstract

SummaryReversed sexual dimorphism in size (RSD) occurs in most species of several taxonomic groups of birds. The hypotheses proposed to explain this phenomenon are examined theoretically, using inequalities to state selection in the most rigorous possible terms. The most pertinent empirical evidence is also examined critically.Proponents of hypotheses on the evolution of RSD have failed to consider the genetic constraints on the evolution of dimorphism. Selection for dimorphism can act on only that small portion of the genetic determination of body size that is sex limited. In general, selection for body size is much more likely to lead to a similar change (e.g. larger) in both sexes than to dimorphism.The most popular hypotheses involve selection for size‐related differences in foraging ability. It is unlikely that there is variation in size‐related foraging differences available for selection in a monomorphic, ancestral population. Foraging differences between the sexes cannot lead to the evolution of RSD; evolution of large and small morphs of both sexes is a more likely outcome. Selection for sex‐role differentiation factors (e.g. large females lay larger eggs, small males are more agile in flight) can lead to the evolution of RSD, but only if the magnitudes of opposing selection for small males and for large females are equal. Combining selection for size‐related foraging differences with selection for sex‐role differentiation factors hinders the evolution of RSD until the sexes differ in size by 3 s.d. Empirical evidence supports this assertion: statistically significant differences between the sexes in the size of prey taken are found only in highly dimorphic species. The sex‐role differentiation factors that have been proposed appear unlikely to provide the equal selection necessary for the evolution of RSD. Several authors have proposed that small size in males is selected for foraging ability and large size in females for some sex‐role differentiation factor. Males cannot be more efficient foragers without females being less efficient and efficiency cannot be a factor only when the male is feeding his family. RSD cannot evolve in monogamous species if large females survive less well than small males.RSD might evolve as the result of sexual selection for small size in males and constraints on the reduction of size in females because of some factor associated with reproduction. Examination of seven studies indicating a relationship between female size and reproductive success shows very little unequivocal evidence for small size in females allowing breeding earlier in the season. Large size in females allows females to breed at a younger age in the sparrowhawk and pairs to form more rapidly in three species of sandpipers. Both of these may be the result of sexual selection.There are fewer theoretical problems with sexual selection as a cause for the evolution of RSD than with the other hypotheses. Empirical evidence for sexual selection is scarce but better than that for the other hypotheses. Evidence is contradictory for the selection of small size in males for agility in aerial displays for courtship or defence of territory. Large size in females does not appear to be the result of selection for competitive ability to obtain mates. Facilitation of female dominance and hence of the formation and maintenance of a pair bond is the most viable explanation of the evolution of RSD. It is most likely that all dimorphism (normal or reversed) is the result of sexual selection.RSD is correlated with birds in the diet in the Falconiformes and this is a central theme in the foraging hypotheses. This correlation may be because birds are abundant and available in a continuum of sizes, thus permitting but not causing the evolution of RSD or because species that prey upon birds are better equipped physically (and perhaps more likely behaviourally) to inflict damaging attacks on conspecifics and the greater RSD increases female dominance and the ease of pair formation.