OBJECTIVES: To investigate the mortality of workers who had been exposed to asbestos, machining fluids and foundry work in a foundry and heavy engineering plant in the railway rolling stock manufacturing industry in New Zealand. METHODS: Historical cohort study design. RESULTS: For the total workforce of 3522 men employed between 1945 and 1991, follow up was 90% of person-years to 31 December 1991. Significantly increased standardised mortality ratios (SMRs) were found for all causes of death combined (SMR 1.07; 95% confidence interval (95% CI) 1.01 to 1.14), all malignancies (SMR 1.15; 95% CI 1.01 to 1.31), circulatory (SMR 1.16; 95% CI 1.07 to 1.27) and musculoskeletal diseases (SMR 3.06; 95% CI 1.39 to 5.84), all digestive cancers (SMR 1.29; 95% CI 1.04 to 1.59), all respiratory cancers (SMR 1.34; 95% CI 1.08 to 1.65), cancer of the oesophagus (SMR 1.97; 95% CI 1.01 to 3.45), and mesothelioma of the pleura (SMR 6.58; 95% CI 1.24 to 19.49). Three deaths from pleural mesothelioma were recorded, with latency times of 51, 53, and 57 years. There were no dose-response relations between exposure to asbestos, machining fluids or foundry work, or by duration of employment in the plant, and any cause of death. CONCLUSIONS: This study found small increases in risk for several causes of death among foundry and heavy engineering workers; however, these increases were small and the possible effects of smoking and other lifestyle factors could not be excluded. There was evidence of asbestos related disease in those involved in engineering work in the past.