Benzyl chloride was administered to groups of 52 male and 52 female F344 rats and (C57BL/6J × BALB/c)F₁ mice at two dose levels by gavage in corn oil three times a week for 2 years. Survivors were sacrificed a few weeks later and examined histopathologically. On the basis of a subchronic study at a range of doses, the doses of benzyl chloride in the chronic study were 100 and 50 mg/kg body weight for mice and 30 and 15 mg/kg body weight for rats. In mice of both sexes there was a high and statistically significant incidence of carcinomas and papillomas of the forestomach. In the livers of male mice, but not of females, there was a significantly increased incidence of hepatocellular neoplasms at the low dose but not at the maximally tolerated dose (MTD). In rats the only neoplasms showing a statistically significant increase compared with controls were C-cell neoplasms of the thyroid gland in females. A significant incidence of neoplasms was not found in the forestomachs of F344 rats, but it is possible that true MTD was not used for rats.