Because large numbers of persons were employed in United States shipyards during World War II, the long-term risks for cancer associated with asbestos exposure in this setting are of great concern. We report here on the mortality findings after up to 29 years of follow-up on a retrospective cohort of 7971 male Pearl Harbor Naval Shipyard workers, which included more than 3000 men whose employment period spanned the World War II years. Compared with the general population of Hawaii, workers in the shipyard cohort had no increase in total mortality or in total cancer mortality irrespective of the duration of their exposure. However, the risk ratio for lung cancer among workers with at least 15 years of asbestos exposure was 1.4 overall (95% confidence interval, 1.0 to 2.0) and 1.7 for those with a latency interval of 30 or more years (95% confidence interval, 1.0 to 2.5). In addition, seven mesotheliomas occurred between 1977 and 1982 in a subset of the cohort, consisting of 7029 Hawaii residents who are being followed prospectively for cancer incidence. This represented an incidence of 67.3 per million men per year, compared with a rate of 5.8 for the state as a whole. These results suggest that the long-term relative increase in risk for mesothelioma may be even greater than that for bronchogenic carcinoma in this and other cohorts of United States shipyard workers exposed to asbestos.