Stomach cancer risk was assessed for smoking and occupation by a population based case/control study. The 739 stomach cancer cases and 3,750 colon cancer cases used as population controls were selected from the Occupational Cancer Incidence Surveillance Study (OCISS) of the Detroit, Michigan metropolitan area. Work, smoking, medical and residential history and demographic information was collected by telephone interviews. Multiple logistic regression analyses were used to calculate maximum likelihood estimates of odds ratios (OR). A significant increase in stomach cancer, with an OR of 1.5, was seen in subjects who had ever smoked cigarettes. The ORs increased with increasing pack/years of smoking. The ORs for men and women smokers were similar; however, among smokers black men had an OR of 2.0 and white women had an OR of 1.7. White male smokers had an OR of 1.2, and black female smokers an OR of 1.4 for stomach cancer. Significant increases of risk were seen, among men, for farming, OR of 2.2; driver sales, OR of 3.8; production inspectors, OR of 2.1; tool and die workers, OR of 1.6; and material workers, OR of 1.9. Significantly increased risks were seen only for white male assemblers, OR of 2.0; mechanics, OR of 2.2; and machine repairmen, OR of 2.1. A significant increase in risk was also observed for white males with more than 10 years employment in the farming industry, with an OR of 2.4. Significant increases in risk were seen for black female assemblers, OR of 5.4; and for white female foodworkers, OR of 4.0. No increases in risk were seen in occupations involving dust exposure. The authors conclude that stomach cancer was associated with cigarette smoking and certain occupations.