The association between employment in motor exhaust-related occupations and the risk for lung cancer was examined in 2,291 male cases of lung cancer and 2,570 controls in data pooled from three U.S. case control studies carried out by the National Cancer Institute between 1976 and 1983. Most analyses were limited to subjects providing direct, in-person interviews, including 1,444 cases and 1,893 controls. For those providing direct interviews and employed 10 years or more in motor exhaust-related (MER) occupations, the age, smoking, and study area adjusted odds ratio (OR) for lung cancer was 1.5 (95% CI = 1.2-1.9). Risk was elevated for truck drivers (OR = 1.5; 95% CI = 1.1-1.9) and for other MER occupations (OR = 1.4; 95% CI = 1.1-2.0). The odds ratios associated with MER employment of 10+ years were 1.6 (95% CI = 1.2-2.1) for whites and 1.4 (95% CI = 0.9-2.1) for nonwhites; 1.2 (95% CI = 0.7-2.0) [corrected] for those with possible exposure to other recognized or reported lung carcinogens; and 1.6 (95% CI 1.2-2.1) for those without such exposure. The 50% excess risk for lung cancer associated with employment in motor exhaust-related occupations could not be explained by greater use of cigarettes or by other occupational exposures among these workers.