A retrospective cohort study was conducted in nine chromeplating plants to examine the mortality of workers employed for at least one year during the period January 1951-December 1981. The study group totaled 178 individuals, 116 of whom were from "hard" and 62 from "bright" chromeplating plants. Vital status ascertainment was 97% complete. The total number of deaths was fairly close to the expected figure (15 observed, 15.2 expected), whereas deaths from tumors exceeded the expected number (8 observed, 4.2 expected). Individuals were distinguished into two subcohorts depending on the exposure intensity, which was much higher in hard than in bright chromeplating. Most deaths from cancer occurred among hard chromium platers, the excess against the expected rate being statistically significant (7 observed, 2.7 expected, p = 0.02). All deaths from lung cancer occurred in this subcohort (3 observed, 0.7 expected, p = 0.03). The increased mortality from cancer among chromium platers seems to be related to exposure intensity and strongly suggests the need for further studies on larger cohorts to confirm the carcinogenicity of chromic acid in man.