A case-control study of 177 premenopausal and 216 postmenopausal breast cancer patients diagnosed in the southern Swedish Health Care Region between 1981 and 1984, and 195 premenopausal and 254 postmenopausal controls randomly chosen during 1984 from the same population by means of a computerized population register was used to estimate the relations among alcohol consumption, cigarette smoking, and the risk of cancer. Cigarette smoking did not seem to be related to the risk for breast cancer among pre- or postmenopausal women. Beer consumption did not show any consistent relation to the risk of breast cancer. An occasional (less frequent than once a week) or a weekly consumption of wine and spirits appeared to be protective compared with abstaining (RRadj = 0.4, 95% CI = 0.3-0.7 for weekly wine consumption, and RRadj = 0.6, 95% CI = 0.4-0.9 for occasional consumption of spirits) among postmenopausal but not among premenopausal women (RRadj = 0.8, 95% CI = 0.7-2.3 and RRadj = 1.0, 95% CI = 0.7-1.8 respectively). The risk estimates were adjusted for differences between cases and controls in age at menarche, age at first full-term pregnancy, age at diagnosis, smoking, and use of exogenous hormones, and for postmenopausal women, age at menopause. The study was based on observations of women with a relatively low exposure to cigarette smoking and alcohol. More frequent exposure or exposure to greater quantities of alcohol and smoking may exhibit different relationships with breast cancer risk.