As part of a large-scale investigation of colorectal cancer incidence, etiology, and survival, a case-control study was conducted to identify whether diet and alcohol, among other variables, were associated with the risk of colorectal cancer. This study compared 715 cases with 727 age- and sex-matched community controls. Findings from the dietary data are presented in the previous paper (Nutr Cancer 9, 21-42, 1987). The total life intake of specific alcoholic beverages was obtained from each subject. Data were classified by consumption of beer, wine, spirits, and alcohol. There was little evidence of an association of any of the alcohol variables with the risk of colon cancer. However, beer was found to be a risk factor for rectal cancer. This effect was more marked in males than in females, but the relative risks for females were consistent with those for males. Relative risk estimates changed only slightly when adjusted for the other alcohol variables and for the variables in the diet model; this suggests that the beer effect is separate from that of other alcohol variables and also from dietary variables. The age differences among beer consumers were found to be associated with cancer risk. Consumption of spirits was associated with a low risk for male rectal cancer. The risk of rectal cancer appeared to depend on beer drinking patterns in the previous 15-20 years.