The relationship between sociodemographic characteristics, lifestyle, family history of cancer, medical history, and reproductive factors and breast cancer was investigated in a population-based case-control study of French Canadians in Montreal. In this study, a total of 414 French-Canadian cases and 429 age- and language-matched population controls were interviewed. Ever-married women showed significantly lower risk (OR: 0.64 [0.45-0.92]) for breast cancer, as did smokers (OR: 0.73 [0.55-0.98]), particularly of nonfilter cigarettes (OR: 0.36 [0.17-0.72]). Weight history, both for the year before the diagnosis of breast cancer and 10 years previously, was associated with risk for the disease. A strong inverse relationship was found between the number of full-term pregnancies (OR: 0.48 [0.28-0.82]) and the risk of breast cancer, while the p trend for late age at first pregnancy (p = 0.02) and menopause (p = 0.004) was statistically significant. A history of breast problems (OR: 1.87 [1.34-2.60]) and a history of breast cancer in relatives (OR: 2.95 [1.63-5.34]) were strongly associated with risk. This study confirms the risk factors of late age at first full-term pregnancy, nulliparity, late age at menopause, and positive family history of breast cancer in the etiology of this disease. Perhaps the protective effect of smoking against breast cancer could be due to its antiestrogenic influence.