In the period between 1980 and 1983 smoking habits, alcohol consumption, coffee drinking, and familial occurrences of peptic ulcers were studied in patients with gastric or duodenal ulcer in North Norway. The results were compared with those in two control groups matched for sex and age but without anamnestically known previous peptic ulcer disease. Statistically significant increased familial occurrences of peptic ulcer were found in relatives of patients both with gastric and with duodenal ulcer, compared with the control group. Furthermore, significantly more smokers were found in the two groups of patients than in the control groups. Patients with duodenal ulcers smoked more than those with gastric ulcer. Both the consumption of coffee and that of alcohol, however, were significantly reduced in the ulcer patients compared with their controls. Therefore, both familial factors and smoking habits appear to have some relationship or even play etiologic roles in the development of peptic ulcer disease, at least in the northern part of Norway. Coffee drinking and intake of alcohol seem to be of no importance. Both tobacco smoking and familial accumulation of peptic ulcers increased the relative risks of getting both gastric and duodenal ulcer as compared with 'non-exposed' persons. Furthermore, a positive correlation was found in men between the quantity of smoking and the risk of developing duodenal ulcer.