The results of epidemiologic studies of the association between cigarette smoking and breast cancer risk have been inconsistent. In spite of the inconsistency, several recent analyses have suggested an increased risk of breast cancer among women who smoked cigarettes for a long period of time and/or who started smoking before their first pregnancy. Our analyses were conducted in the Canadian National Breast Screening Study (NBSS), a multi-center, randomized controlled trial of mammographic screening for breast cancer among 89,835 women aged 40-59 at enrollment. Participants were recruited between 1980 and 1985 from the general Canadian population. During an average of 16.1 years of follow-up, we identified 4,445 incident breast cancer cases. We used the Cox proportional hazards models to estimate multivariate rate ratios (RRs) and 95% confidence limits (CLs) for the association between cigarette smoking and breast cancer. We found that breast cancer risk was associated with the duration (40 years versus 0: RR = 1.50, 95% CL = 1.19, 1.89), intensity (40 cigarettes per day versus 0: RR = 1.20, 95% CL = 1.00, 1.44), cumulative exposure (40 pack-years versus 0: RR = 1.17, 95% CL = 1.02, 1.34), and latency of cigarette smoking (40 years since commencement of smoking versus 0: RR = 1.28, 95% CL = 1.06, 1.55), as well as smoking initiation before a first full-term pregnancy (among parous women, more than 5 years of smoking versus 0: RR = 1.13, 95% CL = 1.01-1.25). These results strongly suggest that cigarette smoking might play an important role in the etiology of breast cancer, particularly when initiated relatively early in life and when engaged in for long durations.