Cancer morbidity and causes of death were examined in a retrospective cohort study of 14,313 male members of the Danish Brewery Workers' Union in order to determine health effects of heavy beer drinking, in particular in the aetiology of colorectal cancer. Brewery workers hold the right to consume 6 bottles (2,100 ml) of light pilsener beer (alcohol content 3.7 g1/00 ml) on the premises of the brewery per working day. The cohort included 1,063 mineral‐water factory workers with no free ration of beer. Cancer morbidity and mortality were compared with those of the general population after adjustment for age, sex, area and time trends. Cancer morbidity, 1943–1972, was increased for cancer of the pharynx (RR = 2.09), oesophagus (RR=2.09), liver (RR=1.51) and larynx (RR = 1.98). The risk of these tumours was highest among workers who had had a ration of free beer during 30 or more years of employment. An increased risk of lung cancer (RR‐1.16) corresponds with the risk among persons of low socio‐economic class. These tumour sites accounted for a 9% excess of all malignant neoplasms among members of the BWU. No increased risk for either cancer of the colon (RR = 1.07) or cancer of the rectum (RR = 1.02) was observed. In contrast to studies on “alcoholics”, only deaths from the above‐mentioned cancers, liver cirrhosis (RR=1.77) and motor vehicle accidents (RR = 1.33) were in excess; total mortality was only slightly above expectation (RR = 1.06). The present investigation suggests that the statistical association between beer and colo‐rectal cancer is of a non‐causal nature. It is concluded that the heavy consumption of “weak” alcoholic beverages may increase the risk of upper aero‐digestive tract cancers, and that part of the disease pattern associated with “alcoholism” is unrelated to alcohol consumption. Copyright © 1979 Wiley‐Liss, Inc., A Wiley Company