LAUSR

Abstract. Behavioral and genetic data were used to determine whether more than one female lays eggs in nests of the Cattle Egret (Bubulcus ibis). Eighty-five nests were monitored, and 18 nests with eggs suspected to be from different females were selected. Both types of eggs were swabbed to sample for DNA since eggshells contain cells from the reproductive tract of the mother or the skin of incubating individuals. Using these swabs, molecular sexing was first conducted, and three clutches were discarded due to samples with a pattern of male DNA. In the remaining 15 nests, DNA was extracted from swabs (n = 45 eggs) and genotyped at seven microsatellite loci. The genetic diagnosis was developed with a blinded protocol including both suspected and non-suspected eggs by comparison of maternal DNA found on the eggs. With one exception, genotypes not compatible with the predominant female genotype in each clutch were found almost exclusively on suspected eggs, proving that the genetic diagnosis was not an artifact. Using a behavioral criterion, multiple laying females were detected in 21.1% nests, but genetic data showed that the behavior methodology should be modified. In the 15 nests analyzed, the behavioral criterion failed to detect one suspected egg in one nest ( #11) (false negative) and indicated two suspected eggs in two nests (#12 and #13) that were not confirmed using the genetic criterion (false positives). The results showed that estimates based solely on the behavior approach can be biased, and the genetic method should be applied to check and improve ecological data.