The risk of suicide is well known to be increased among heavy alcohol drinkers. However, whether the risk is increased or decreased among light drinkers is still under debate. We investigated this association in a population-based sample of men in Japan. The Ohsaki Study was a population-based, prospective cohort study among Japanese adults aged from 40 to 79 years. Between October and December, 1994, 22,804 men in Miyagi Prefecture, Japan, completed a questionnaire on various health-related lifestyles, including alcohol drinking. During the subsequent 7 years follow-up, 73 participants committed suicide. We used the Cox proportional hazards regression model to estimate the hazard ratio (HR) for suicide mortality according to the quantity of alcohol consumed daily, with adjustment for potential confounders. There was a statistically significant positive and linear association between the amount of alcohol consumed and the risk of suicide: the multivariate HRs in reference to nondrinkers (95% confidence interval) were 1.2 (0.5-2.7), 1.5 (0.7-3.4), and 2.4 (1.2-4.6) in current drinkers who consumed </=22.7g, 22.8g-45.5g, and >/=45.6g of alcohol per day, respectively (P-trend=.016). Even after the early death cases were excluded, a significant linear association was observed between alcohol consumption and the risk of suicide, with the risk of suicide also being nonsignificantly higher among the light drinkers than among nondrinkers (multivariate HR=1.7). This prospective cohort study indicated a positive linear association between alcohol consumption and the risk of suicide, and the suicide risk among the light drinkers was not decreased as compared with that in nondrinkers.