The loggerhead turtle (Caretta caretta) is listed as an endangered species in the International Union for Conservation of Nature Red List, and this is critical for the maintenance of future genetic diversity. For the conservation of this species, it is important to understand the relationships between sexual selection and gene flow in the evolution of biodiversity. It is, however, virtually impossible to study sexual selection in such an oceanic species in the field. We investigated the relationships between mate preference and genetic relatedness of reproductively active loggerhead turtles (n  =  7) kept in a tank at an aquarium using 4 different estimators of pairwise genetic relatedness (r) based on the genotypes at 23 microsatellite loci. Although relationships between total number of courtship behavior and r were not significant in the 4 estimators, there were significant and consistent inverse relationships between cumulative duration of mountings and r in all of the estimators. In addition, significant or marginally significant relationships were found between mean r for each female and number of sires in her successive clutches in 3 of the 4 estimators. However, we found no evidence that more distantly related pairs produced more offspring than the assumption of random mating except for 1 estimator or any relationship between r of parents and their reproductive success (hatchability). Based on these results, the most parsimonious explanation is that loggerhead turtles in the tank tend to prefer their mate(s) according to genetic relatedness but that the extent of female mate selection and the relationship between mate selection and fitness still remain debatable.