A total of 1,162 prostate cancer cases and 3,124 age-matched hospital controls from several U.S. hospitals were studied to identify associations between prostate cancer and life-style variables. Among white males, college education, professional occupation, and non-Jewish ethnicity were weakly associated with the risk of prostate cancer (ORs = 1.5, 1.8, and 1.3, respectively, P less than 0.01). These relationships retained statistical significance after adjustment for age, marital status, body mass, cigarette smoking, alcohol consumption, and physical activity in multiple logistic regression models. Among black males, similar, though nonsignificant, risk elevations were observed for education and occupation. Weak positive effects of borderline statistical significance were observed for high body mass, low physical exercise, and high serum cholesterol (in elderly males only), and a significant reduction in risk was noted for never married black males. Cigarette smoking and alcohol consumption were not related to the risk of prostate cancer.