This study evaluated cancer incidence and mortality among 251 employees of a facility that synthesized a benzothiadiazin herbicide between 1979-1987. It was initiated in response to a perceived cancer cluster. Cancers were identified mainly by review of occupational medical records and personal interviews. Standardized incidence ratios (SIRs) were computed using comparison data provided by the Louisiana Tumor Registry. Overall mortality was less than expected and did not vary by job type or duration of assignment in the facility. Total cancer incidence was marginally elevated [SIR = 1.4; 95% Confidence Interval (CI): 0.9-1.9] and was highest among men assigned to the unit during 1979 [SIR = 2.1; 95% CI: 1.3-3.3]. Thirteen of 20 cancers among these men were either prostate (9) or digestive system (4) cancers. Increased medical examinations and cancer awareness may have contributed to the detection of cancers at earlier ages than would be seen in a general population. No specific workplace agents were identified to account for the findings although a workplace role cannot be ruled out.