The rodent liver carcinogen, mitogen, and peroxisome proliferator, methylclofenapate (MCP), has been investigated as a chemical mitogen for use in the rat liver micronucleus assay. A series of experiments comparing MCP with the potent mitogen 4-acetylaminofluorene (4AAF) and with 2/3 partial hepatectomy showed unexpected differences between the three treatment regimes as monitored by the expression of micronucleated hepatocytes (MH) in livers initiated with N-nitrosodimethylamine (NDMA). These differences were observed in both the profile of MH induction, as a function of time after mitogenic stimulation, and in the magnitude of response. There was no correlation between the magnitude of MH induction and the degree of mitogenesis triggering the MH induction. It is concluded that the yield of MH observed in a rat liver micronucleus assay is not a simple function of the level of DNA damage induced by the initiating agent (here NDMA). This indicates the need for caution in the interpretation of data obtained in the liver micronucleus assay for chemicals of unknown carcinogenicity. The use of acridine orange as a fluorescent stain for use with isolated hepatocytes is described.