In 1980, 87,526 female nurses 34 to 59 years of age completed a dietary questionnaire that assessed their consumption of beer, wine, and liquor. By 1984, during 334,382 person-years of follow-up, we had documented 200 incident cases of severe coronary heart disease (164 nonfatal myocardial infarctions and 36 deaths due to coronary disease), 66 ischemic strokes, and 28 subarachnoid hemorrhages. Follow-up was 98 percent complete. As compared with nondrinkers, women who consumed 5 to 14 g of alcohol per day (three to nine drinks per week) had a relative risk of coronary disease of 0.6 (95 percent confidence interval, 0.4 to 0.9); for 15 to 24 g per day the relative risk was 0.6 (0.3 to 1.1), and for 25 g or more per day it was 0.4 (0.2 to 0.8), after adjustment for risk factors for coronary disease. Alcohol intake was also associated with a decreased risk of ischemic stroke. For 5 to 14 g of alcohol per day the relative risk was 0.3 (0.1 to 0.7), and for 15 g per day or more it was 0.5 (0.2 to 1.1). In contrast, although the number of cases of subarachnoid hemorrhage was small, alcohol intake tended to be associated with an increased risk of this disorder; for 5 to 14 g per day the relative risk was 3.7 (1.0 to 13.8). These prospective data suggest that among middle-aged women, moderate alcohol consumption decreases the risks of coronary heart disease and ischemic stroke but may increase the risk of subarachnoid hemorrhage.