The objective of this study was to investigate the induction of liver tumors by arylhydroxamic acids. The potential involvement of sulfate conjugation was minimized by the administration of a N-hydroxy-4-acylaminobiphenyl to female CD rats. This experimental design provided for the exposure of a target organ that has only a low capacity for activation of hydroxamic acids by sulfate conjugation, with a carcinogen that does not induce tumors in livers that possess a high sulfotransferase activity. A single dose of the N-formyl or N-acetyl derivatives of N-hydroxy-4-aminobiphenyl was given i.p. at 0.4 mmol/kg body weight to 34-day-old animals. In attempts to amplify the hepatocarcinogenic potential of the compounds, partial hepatectomy 24 hr before the chemical injection and subsequent long-term treatment with phenobarbital in the diet were carried out. For comparative purposes, other animals were subjected to three additional partial hepatectomies subsequent to the carcinogen administration instead of the phenobarbital treatment. The experiments were terminated 64 weeks after injection. Both the N-formyl and N-acetyl derivatives of N-hydroxy-4-aminobiphenyl, in conjunction with partial hepatectomy and subsequent treatment of dietary phenobarbital, induced a high incidence of neoplastic nodules and γ-glutamyltranspeptidasepositive foci in the liver. Only one hepatocellular carcinoma was observed in each treatment group. Repeated partial hepatectomies enhanced the yield of γ-glutamyltranspeptidasepositive foci but were ineffective in producing neoplastic nodules. In addition to the liver lesions, mammary tumors were also induced. Importantly, an inhibitory effect of the subsequent administration of phenobarbital was observed on mammary tumor formation, possibly because of alterations in hormone metabolism resulting from the induction of microsomal enzymes by phenobarbital, which resulted in a decreased promoting effect. There was no difference in the tumorigenicity of the formyl and acetyl derivatives in these experiments.