Brood sex ratio varies with diet composition in a generalist raptor

While sex allocation has been investigated productively at both population and family levels, as yet no general theory has been developed that is capable of linking processes at these two ecological scales, and very few empirical studies have examined cross‐scale patterns. In Finnish northern goshawks (Accipiter gentilis), nestling sex ratio of local subpopulations is related to the spatial and temporal variation in the abundance of their principal avian prey, woodland grouse. Using data from an urban breeding population in Hamburg, Germany, I investigated: (1) whether brood sex ratio of goshawks varies with diet composition at the family level; (2) whether such variation could reflect adaptive adjustment; and (3) how family‐level allocation can drive population‐level patterns, such as those observed in Finland. Feral pigeons (Columba livia) were the most important prey species, with a pooled contribution to total diet of 36%. Brood sex ratio varied significantly with the proportion of pigeons in the breeding‐season diet of pairs (increasing male bias). However, there was no evidence for sex‐differential effects of diet composition, so it remains unclear whether the observed sex‐ratio variation was an adaptive response. As all study pairs inhabited an (urban) environment where pigeons were unusually abundant, family‐level sex‐ratio adjustment caused a marked male bias in offspring sex ratio at the population level (male‐biased nestling sex ratio in four of five years; pooled data: 60% males). This suggests that the large‐scale variation observed in Finnish goshawk populations mirrors sex‐ratio adjustment shown by individual families in response to small‐scale environmental conditions. Apart from linking patterns empirically across ecological scales, this study is, to my knowledge, the first to demonstrate that family‐level brood sex ratio varies with realized resource use (diet composition) in a raptor species. Previous studies either failed to find significant associations or, more commonly, violated theoretical assumptions by measuring environmental prey abundance (often integrated over large areas) rather than realized prey use of individual breeding pairs. I conducted a meta‐analysis of offspring sex‐ratio data from 17 goshawk populations across Europe to put my results into perspective. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 105 , 937–951.