The function of female birds' extra-pair (EP) behavior has remained an unresolved question in ornithology and behavioral ecology for >30 years. The genetic compatibility hypothesis (GCH) proposes that females benefit by acquiring biological sires that yield more heterozygous, and therefore fitter, offspring than their social mates. We used ten polymorphic microsatellite loci to test GCH predictions and its assumption that fitness increases with heterozygosity in blue-footed boobies (Sula nebouxii), a long-lived tropical seabird. Our predictions were not supported. Heterozygosity was uncorrelated with quality indicators (fledging probability, fledgling or adult body size or mass, adult ornamentation, mean breeding success). Females were no more likely to have EP behavior or chicks when their social mates were less heterozygous or compatible, nor were EP males more heterozygous or compatible than the males they cuckolded. Finally, EP chicks were no more heterozygous than within-pair chicks overall or in half-sib comparisons, nor were within-pair chicks from all-within-pair nests more heterozygous that those with EP nestmates. There are both methodological and biological explanations for these consistently negative results. Inadequate sample size is possible but unlikely, since our samples were comparable or larger than those of similar studies with significant findings. Lack of identity disequilibrium (within-individual heterozygosity correlation) among our marker loci could be responsible, and suggests either insufficient marker coverage or lack of inbreeding, bottleneck, and/or admixture. An independent social pedigree revealed infrequent inbreeding, suggesting that pressure on females to select sires that maximize offspring heterozygosity may be genuinely lax. Alternatively, it is possible that the GCH is only upheld when selection on young is strongest; this would not be detected in our sample, which was taken during an extremely productive year. Whatever their cause, our results
Accepted Ar ticle'This article is protected by copyright. All rights reserved.' expand the taxonomic breadth of avian EP behavior analyses and should be considered in future evaluations of the GCH.