In recent years it has become increasingly evident that the toxicity produced by a variety of compounds can be attributed to their metabolites. Simple dose-toxicity studies of a metabolite will help to elucidate its toxic effect, but it is not possible to quantify its role in the toxicity produced by the parent compound unless the disposition of the preformed and endogenously formed metabolite is taken into account. We assessed the contribution of the metabolite, R-(+)-menthofuran (MF), to the hepatotoxicity observed after i.p. administration of R-(+)-pulegone (PUL) to rats. As the major constituent of pennyroyal oil, PUL has been linked to the deaths of several young women over the past several years. After i.p. administration of PUL and MF to separate groups of rats at doses selected to match the area under the curve of generated and synthetic MF, plasma alanine transferase was comparable between groups. Although the area under the curve was matched, the peak concentration in plasma of MF formed in vivo is 4.5 times higher and occurs much earlier than after the administration of MF itself. When the exposure of rats to preformed and synthetic MF is matched with respect to time course in plasma, PUL produces more than twice the increase in plasma alanine transferase and hepatocellular necrosis than does MF. The results suggest that events other than those associated with the disposition of MF contribute to the hepatotoxicity observed after ingestion of PUL, and they emphasize the importance of time course in assessing the role of a metabolite in the toxicity of a given compound.