An evolutionary explanation for seasonal trends in avian sex ratios

We present an extensive set of data for five species of raptorial birds to demonstrate that some raptor species produce an excess of daughters early in the season and an excess of sons in late nests, while others show the reverse. By means of a simulation model we investigate an evolutionary explanation for this phenomenon in terms of sex-specific differences in the relation between age at first breeding and date of birth. The model predicts diat that gender should be produced first in the season whose age of first breeding is more strongly accelerated by an early birth date. We argue that this tends to be the male gender in raptor species, such as the common kestrel (Falco tinnunculus), which tend to breed early in life, while it is the female gender in larger species with later onset of breeding, such as the marsh harrier (Circus aeruginosus). The empirical evidence is qualitatively consistent with this hypothesis. Our model is quite general in that it makes no assumptions about the mechanism (primary sex-ratio bias at egg laying or secondary sex-differential mortality before fledging) by which the bias is generated. Yet it is able to create quantitative predictions for species where sufficient demographic and life-history data are available. From die available data set in the common kestrel we derive a quantitative prediction for the seasonal trend in brood sex ratio. The observed trend is in good agreement with this prediction.