OBJECTIVES: To evaluate the mortality of a group of tannery workers. METHODS: The cohort consisted of 1244 workers (870 men and 374 women) employed at a chrome tannery between 1955 and 1988. A total of 36414 person-years of follow up was calculated (369 people had died). National and regional mortalities were used to estimate the expected numbers. RESULTS: All cause mortality was similar to that of the general population. The most remarkable excess was for bladder cancer (observed 10, standardised mortality ratio (SMR) 242, 95% confidence interval (95% CI) 116 to 446). An excess of colorectal cancer (observed 17, SMR 180, 95% CI 105 to 288) was also found, based on an increased risk of both colon (SMR 166) and rectal cancer (SMR 206). No recognisable patterns emerged from the analyses by years since first employment, calendar year of hire, or lagging exposures. CONCLUSIONS: The increased mortality from bladder cancer is likely due to exposure to benzidine based leather dyes. If the apparent excess of colorectal cancer is real, its causes are as yet unknown.