A population-based sample of 1673 (valid phenotypic and genotypic data were available from 1412 individuals) Finnish male twins and siblings of twins aged 18-45 years provided questionnaire data regarding ejaculatory function as well as saliva samples for genotype analyses. Genotypic analyses were conducted controlling for between-subjects dependence. No significant association was found between the 5-HTTLPR polymorphism and a composite variable measuring premature ejaculation or between this polymorphism and a self-report measure of ejaculation latency time. Previously conducted studies have found contradicting results regarding the possible role of 5-HTTLPR in premature ejaculation. Methodological inconsistencies have been pointed out in these studies, which have all been conducted on rather small samples. While differences in terms of measurement of ejaculatory function could partly explain why positive findings from some earlier studies could not be replicated, the present study, given the large sample size and multifactorial measures used, indicated that the 5-HTTLPR polymorphism has a limited, if any, impact on ejaculatory function.