Male F344/N rats were dosed with ethyl acrylate (EA) either by daily gavage or in the drinking water for 2 weeks. The gavage dose levels were 0, 2, 10, 20, 50, 100, and 200 mg/kg; the drinking water dose concentrations were 0, 200, 1,000, 2,000, and 4,000 ppm (corresponding to 0, 23, 99, 197, and 369 mg/kg/day, respectively). In those animals dosed by gavage, irritation of the forestomach increased in incidence and severity over the 20-200 mg/kg dose range. In those animals dosed with EA in the drinking water, a much lower incidence of forestomach irritation and less severe lesions were observed at corresponding dose levels. No lesions were observed in the glandular stomach from either of the 2 modes of oral administration. Following 2 weeks of gavage dosing with EA, the total non-protein sulfhydryl (NPSH) content of the forestomach and glandular stomach, and the NPSH concentration of the liver were determined 2-24 hr after the last gavage dose. Animals dosed at 200 mg/kg reached approximately 11% of the initial NPSH content in the forestomach at 6 hr after dosing. NPSH depletion of this magnitude has been associated with cytotoxicity of other tissues in other studies. By contrast, either the glandular stomach nor liver were depleted of NPSH to levels generally associated with toxicity. These observations are consistent with the conclusion that bolus dosing of EA induces severe depletion of critical cellular thiols in the forestomach with toxic consequences, but not in the glandular stomach or liver. Changing the mode of oral administration for EA to continued small doses in the drinking water allowed efficient detoxification and did not induce sulfhydryl depletion or comparable forestomach toxicity at the same daily body burden.