OBJECTIVES: To examine the hypothesis that, theoretically at least, exposure to toxicants of the type present in the Gulf war could affect spermatogenesis, which might be observed as increased levels of infertility. DESIGN: Retrospective reproductive cohort analysis. SETTING: Male UK Gulf war veterans and matched comparison group of non-deployed servicemen, surveyed by postal questionnaire. PARTICIPANTS: 42,818 completed questionnaires were returned, representing response rates of 53% for Gulf veterans and 42% for non-Gulf veterans; 10,465 Gulf veterans and 7376 non-Gulf veterans reported fathering or trying to father pregnancies after the Gulf war. MAIN OUTCOME MEASURES: Failure to achieve conceptions (type I infertility) or live births (type II infertility) after the Gulf war, having tried for at least a year and consulted a doctor; time to conception among pregnancies fathered by men not reporting fertility problems. RESULTS: Risk of reported infertility was higher among Gulf war veterans than among non-Gulf veterans (odds ratio for type I infertility 1.41, 95% confidence interval 1.05 to 1.89; type II 1.50, 1.18 to 1.89). This small effect was constant over time since the war and was observed whether or not the men had fathered pregnancies before the war. Results were similar when analyses were restricted to clinically confirmed diagnoses. Pregnancies fathered by Gulf veterans not reporting fertility problems also took longer to conceive (odds ratio for > 1 year 1.18, 1.04 to 1.34). CONCLUSIONS: We found some evidence of an association between Gulf war service and reported infertility. Pregnancies fathered by Gulf veterans with no fertility problems also reportedly took longer to conceive.