A general mortality survey was conducted on chemical manufacturing workers. The sample consisted of 2,115 active and former employees engaged in production and nonproduction activities. Cause specific mortality in the sample was compared with that expected based on worker years at risk. Allowance was made for potential latent effects of occupational exposure by performing an analysis that ignored the first 15 years at risk after hire. A second analytical approach was taken to compare the risk of selected causes of death among the various subgroups of the sample defined by duration of employment and period of hire. Nearly half the sample was employed with the company for less than 1 year; two thirds had been employed for less than 5 years. Elevated standard mortality rates were noted for three general cause of death categories: malignant neoplasms; all external causes of death; and symptoms, senility, and ill defined conditions. A significant excess mortality from respiratory malignancies was observed and was attributed primarily to lung cancer. Excess mortality was also observed for cancer of the kidney and pancreas. In workers with over 15 years of service, only cancers of the lung and kidney were significantly in excess. The influence of year of hire, duration of employment, and interval since hire on major cause mortality risks was not significant in a majority of cases. Kidney cancer cases were slightly older at hire than either the lung or pancreatic cancer cases. The authors conclude that there is a significant excess mortality due to cancer among chemical manufacturing workers and is attributed to an excess of cancers of the kidney, lung, and pancreas.