The effect of the type of carcinogen initiator on the ability of phenobarbital to promote hepatic tumor formation in 15-day-old initiated male B6C3F1 mice was evaluated. Fifteen day old male B6C3F1 mice were divided into 6 groups of 10 mice each. Groups 1 and 2 received a single intraperitoneal (ip) injection of diethylnitrosamine (DENA) (5 micro g/gbw). Groups 3 and 4 received a single ip injection of dimethylnitrosamine (DMNA) (5 micro g/gbw). Groups 5 and 6 received a single ip injection of saline. At weaning (28 days of age), mice in groups 2, 4, and 6 received phenobarbital (500 mg/ml) in their drinking water. Mice in groups 1, 3, and 5 received deionized drinking water. Drinking water treatment continued for 24 weeks at which time mice were sampled. At sampling, mice were examined for hepatic tumors by histology. Mice in Groups 5 (no treatment) and 6 (PB only) did not exhibit hepatic tumors. Groups 2 (DENA + PB) displayed a decrease in hepatic adenomas from that of Group 1 (DENA only) confirming previous observations. Treatment with DMNA and PB (Group 4) resulted in a significant increase in both hepatic adenoma incidence and number over that of DMNA only (Group 3) treated mice. (Copyright (c) 1988 Elsevier Scientific Publishers, Ireland Ltd.)