A study of colon and stomach cancer mortality in automotive wood model makers was conducted. The cohort consisted of 2294 white male wood model makers employed for at least 1 month between January 1, 1940 and December 31, 1980 at one factory of each of the three major automobile companies in the Detroit, Michigan area. The vital status of the cohort was determined as of December 31, 1984. Causes of death of the decedents were determined. Standardized mortality ratios (SMRs) were computed using mortality rates of the general white male population of the United States (US) as the reference. SMRs for colon and stomach cancer were also computed using mortality rates of Wayne County, Michigan as the reference. When compared to US mortality rates, overall mortality in the cohort was significantly decreased, SMR 0.8. Mortality from colon and stomach cancer was nonsignificantly elevated, SMRs 1.6 and 1.2, respectively. Mortality from other cancer sites was not significantly elevated. Mortality from most nonmalignant causes of death was significantly decreased. When compared with Wayne County mortality rates, the SMRs for colon and stomach cancer were 0.9 and 1.6, respectively. A nested case/control study of 20 colon cancer and 17 stomach cancer cases and 543 age matched referents that examined exposure to wood dust or length of employment in wood model making as risk factors was conducted. No significant associations between wood dust exposure or length of employment in model making and mortality from colon or stomach cancer were found.