We conducted a follow-up study to evaluate mortality among 14,861 workers employed in five facilities producing or using phenol and formaldehyde. More than 360,000 person-years of follow-up accrued. Mortality rates from all causes of death combined were similar to those in the general U.S. population. We observed excesses of cancer of the esophagus, cancer of the kidney, and Hodgkin's disease among workers exposed to phenol, but none of these excesses showed a dose-response relation with exposure to phenol. Excess lung cancer mortality (SMR = 1.2) showed no consistent pattern by any exposure index. Workers exposed to phenol had lower mortality ratios for cancer of the buccal cavity and pharynx, cancer of the stomach, cancer of the brain, arteriosclerotic heart disease, emphysema, disease of the digestive system, and cirrhosis of the liver. Of these, arteriosclerotic heart disease, emphysema, and cirrhosis of the liver were inversely related to duration of phenol exposure and to cumulative phenol exposure levels. Although these inverse associations may be due to chance or uncontrolled confounders, the ability of phenol to interfere with the generation of oxidants in experimental systems suggests that the pattern may have biologic plausibility.