Mortality in a cohort of 770 workers with potential pentachlorophenol (PCP) exposure was evaluated from 1940 through 1989. The study cohort is a subset of a larger cohort of workers with potential exposure to higher chlorinated dioxins. Total mortality and cancer mortality in the PCP cohort were slightly lower than expected in comparison to the U.S. white male population. There were 229 total deaths with 242.5 expected (SMR = 94, 95% confidence interval 83-108), and 50 cancer deaths with 52.6 expected (SMR = 95, 95% confidence interval 71-125). In comparison with unexposed employees, the risk ratio for total mortality was 1.03 (95% confidence interval 0.90-1.17), and the risk ratio for all cancer mortality was 0.95 (95% confidence interval 0.71-1.26). In most cause of death categories of a priori interest no deaths were observed in the cohort. A small excess of other and unspecified lymphopoietic cancer deaths was observed but did not appear to be related to exposure. Excesses of deaths due to cancer of the kidney, gastric and duodenal ulcer, cirrhosis of the liver, and all accidents were observed in comparison with the U.S. white male population and with unexposed employees. These were associated with increasing estimated cumulative PCP exposure after lagging exposures by 5 and 15 years. Despite the limited size and the generally favorable total mortality experience of the cohort, it was concluded that cohort members may have incurred increased risk of death due to some specific causes. The risks could not, however, be attributed conclusively to PCP exposure and may have been associated with other occupational and nonoccupational factors. Additional mortality surveillance of this cohort will be performed.