We analyzed data from a population-based case-control study of Wisconsin women to evaluate the relationship of alcohol consumption to endometrial cancer risk. Cases (n = 739) were identified from a statewide tumor registry; controls (n = 2313) were selected randomly from driver's license lists and Medicare beneficiary files. Alcohol consumption and other factors were ascertained by telephone interview. Compared with abstainers, the multivariable relative risk for recent consumption of two or more drinks per day was 1.27 [95% confidence interval (CI) 0.78-2.07]; increasing consumption was not associated with risk of disease (P for trend, 0.82). The relative risk for early adulthood consumption of two or more drinks per day was 1.00 (95% CI, 0.58-1.73), with no suggestion of a trend (P = 0.26). Although the sample size was limited, a significant inverse association was suggested in premenopausal women consuming one drink per day or more (0.20, 95% CI 0.06-0.71). Beverage-specific consumption was not associated with risk. This study suggests that, unlike breast cancer, endometrial cancer is not positively associated with alcohol intake.