The promotion-suppressing ability of two antioxidants was measured to determine the role of oxidative stress in hepatocarcinogenesis. Four-day-old female F344/N rats were dosed with diethylnitrosamine (10 mg/kg). After weaning, they were fed semipurified diets with and without 500 ppm alpha-tocopherol, or the same two diets containing 500 ppm phenobarbital, or 5,000 ppm butylated hydroxyanisole (BHA) for 3 or 11 months. By 11 months, phenobarbital-fed groups had eaten 30% more than other groups did (p less than 0.05), suggesting a role for increased caloric intake in phenobarbital promotion. Phenobarbital and BHA significantly reduced body weights and increased liver weights compared with control rats. After three months, alpha-tocopherol significantly suppressed mean volume of placental glutathione S-transferase (PGST)-positive altered hepatic foci (AHF), regardless of xenobiotic treatment. Phenobarbital increased and BHA decreased the numbers of AHF compared with those of the control group. After 11 months, mean focal volume was significantly suppressed by BHA compared with that of the control group, and phenobarbital increased the total volume of AHF [PGST-positive plus gamma-glutamyltransferase (GGT)-positive AHF] compared with rats fed either control or BHA diets. BHA treatment also increased hepatic glutathione levels by 40% compared with control and rats fed phenobarbital. In conclusion, alpha-tocopherol had only a slight, early effect to suppress promotion of hepatocarcinogenesis. BHA suppressed some indices of promotion at both times and increased hepatic glutathione; however, BHA's toxicity (which suppressed body weight) may also be a factor in its supposable promotion-inhibitory effects.