1,3-Butadiene (BD) is a more potent tumor inducer in mice than in rats. BD also shows striking differences in metabolic activation, with substantially higher blood concentrations of 1,2:3,4-diepoxybutane (butadiene diepoxide; BDE) in BD-exposed mice than in similarly exposed rats. The objective of this study was to develop a single mechanistic model structure capable of describing BD disposition in both species. To achieve this objective, known pathways of 1,2-epoxy-3-butene (butadiene monoepoxide; BMO) and BDE metabolism were incorporated into a physiologically based pharmacokinetic model by scaling rates determined in vitro. With this model structure, epoxide clearance was underestimated for both rats and mice. Improved simulation of blood epoxide concentrations was achieved by addition of first-order metabolism in the slowly perfused tissues, verified by simulation of data on the time course for BMO elimination after i.v. injection of BMO. Blood concentrations of BD were accurately predicted for mice and rats exposed by inhalation to constant concentrations of BD. However, if all BD was assumed to be metabolized to BMO, blood concentrations of BMO were overpredicted. By assuming that only a fraction of BD metabolism produces BMO, blood concentrations of BMO could be predicted over a range of BD exposure concentrations for both species. In vitro and in vivo studies suggest an alternative cytochrome P-450-mediated pathway for BD metabolism that does not yield BMO. Including an alternative pathway for BD metabolism in the model also gave accurate predictions of blood BDE concentrations after inhalation of BD. Blood concentrations of BMO and BDE observed in both mice and rats are best explained by the existence of an alternative pathway for BD metabolism which does not produce BMO.