The International Agency for Research on Cancer (IARC) has found that the evidence for the carcinogenicity of beryllium is sufficient based on animal data but "limited" based on human data. This analysis reports on a retrospective cohort mortality study among 9,225 male workers employed at seven beryllium processing facilities for at least 2 days between January 1, 1940, and December 31, 1969. Vital status was ascertained through December 31, 1988. The standardized mortality ratio (SMR) for lung cancer in the total cohort was 1.26 (95% confidence interval [CI] = 1.12-1.42); significant SMRs for lung cancer were observed for two of the oldest plants located in Lorain, Ohio (SMR = 1.69; 95% CI = 1.28-2.19) and Reading, Pennsylvania (SMR = 1.24; 95% CI = 1.03-1.48). For the overall cohort, significantly elevated SMRs were found for "all deaths" (SMR = 1.05; 95% CI = 1.01-1.08), "ischemic heart disease" (SMR = 1.08; 95% CI = 1.01-1.14), "pneumoconiosis and other respiratory diseases" (SMR = 1.48; 95% CI = 1.21-1.80), and "chronic and unspecified nephritis, renal failure, and other renal sclerosis" (SMR = 1.49; 95% CI = 1.00-2.12). Lung cancer SMRs did not increase with longer duration of employment, but did increase with longer latency (time since first exposure). Lung cancer was particularly elevated (SMR = 3.33; 95% CI = 1.66-5.95) among workers at the Lorain plant with a history of (primarily) acute beryllium disease, which is associated with very high beryllium exposure. The lung cancer excess was not restricted to plants operating in the 1940s, when beryllium exposures were known to be extraordinarily high. Elevated lung cancer SMRs were also observed for four of the five plants operating in the 1950s for workers hired during that decade. Neither smoking nor geographic location fully explains the increased lung cancer risk. Occupational exposure to beryllium compounds is the most plausible explanation for the increased risk of lung cancer observed in this study. Continued mortality follow-up of this cohort will provide a more definitive assessment of lung cancer risk at the newer plants and among cohort members hired in the 1950s or later at the older plants. Further clarification of the potential for specific beryllium compounds to induce lung cancer in humans, and the possible contribution of other exposures in specific processes at these plants, would require a nested case-control study. We are currently assessing whether available industrial hygiene data would support such an analysis.