Background: Epidemiologic studies of men consistently demonstrate a positive association between tobacco smoking and risk of colorectal adenomas, precursors of cancer, but have not consistently shown an association between smoking and colorectal cancer. We hypothesized that smoking acts as an initiator of colorectal neoplasia and that the association with cancer has been obscured because the time is long between onset of smoking and diagnosis of cancer. Purpose: Our purpose was to examine the association between cigarette smoking and risk of colorectal adenoma and colorectal cancer in men and to estimate the minimum induction period between the start of smoking and the diagnosis of cancer. Methods: Using data from the ongoing Health Profesionals Follow-up Study, we assessed the relative risk (RR) of small adenoma, large adenoma (≥1cm), and cancer according to pack-years of smoking. Current and lifetime histories of smoking and other confounding factors were assessed by questionnaire at baseline and at 2-years intervals. Between 1986 and 1992, we documented 238 new cases of colorectal cancer among 47935 U.S. males and 626 new cases of colorectal adenomas among 12854 of the men who had a sigmoidoscopy or colonoscopy. Results: Smoking during the prior two decades was associated with the prevalence of small adenomas (RR = 2.96; 95% confidence interval (CI) = 1.47-5.98; for ≥35 pack-years versus 0 pack-years within the 20 years preceding the endoscopy, P trend =.04) but not with large adenomas (RR = 0.46; 95% CI = 0.11-1.94; P trend =.56). However, smoking more than 20 years in the past was associated with large adenomas (RR = 2.38; 95% CI = 1.56-3.63; P trend =.004 for smoking ≥16 pack-years versus 0 pack-years). Smoking was related to risk of colorectal cancer only after allowing for an induction period of at least 35 years (RR = 1.94; 95% CI = 1.13-3.35; P trend =.008 for smoking ≥16 versus 0 pack-years more than 35 years in the past). Conclusions: Smoking in the prior 20 years has a strong relation to small colorectal adenomas, smoking at least 20 years in the past is related to larger adenomas, and the induction period for colorectal cancers is at least 35 years. Implications: Our results highlight the need to intensify efforts to prevent smoking, especially among the young, and suggest a reduced threshold for screening for colorectal cancer among long-term smokers. [J Natl Cancer Inst 86:183-191, 1994].