First-pass metabolism of 1,3-butadiene (BD) leading to 1,2-epoxy-3-butene (EB), 1,2:3,4-diepoxybutane (DEB), 3-butene-1,2-diol (B-diol), 3,4-epoxy-1,2-butanediol (EBD) and crotonaldehyde (CA) was studied quantitatively in the once-through BD perfused liver of mouse and rat by means of an all-glass gas-tight perfusion system. Metabolites were analyzed using gas chromatography equipped with mass selective detection. The perfusate consisted of Krebs-Henseleit buffer (pH 7.4) containing bovine erythrocytes (40%v/v) and BD. The perfusion flow rates through the livers were 3-4 ml/min (mouse) and 17-20 ml/min (rat). The BD concentrations in the liver perfusates were 330 nmol/ml (mouse) and 240 nmol/ml (rat) being high enough to reach almost saturation of BD metabolism. The mean rates of BD transformation were about 0.014 and 0.055 mmol/h per liver of a mouse and a rat, respectively, being similar to the values expected from in-vivo measurements. There were marked species differences in the formation of BD metabolites. In the effluent of mouse livers, all three epoxides (EB: 9.4 nmol/ml; DEB: 0.06 nmol/ml; EBD: 0.07 nmol/ml) and B-diol (8.2 nmol/ml) were detected. In the perfusate leaving naïve rat livers, only EB and B-diol were found. In that of rat liver, EB concentration was 8.5 times smaller than in that of mouse liver, whereas B-diol concentrations were similar in the effluent liver perfusate of both species. CA was below the limit of its detection (60 nmol/l) in the liver perfusate of mice and of naïve rats. Of BD metabolized, the sum of the metabolites investigated in the effluent amounted to only 30% (mouse) and 20% (rat). In first experiments with rat liver, glutathione (GSH) was depleted by pretreating the animals with diethylmaleate. With the exception of EBD (not quantifiable due to an interfering peak), all other metabolites including CA were found in the effluent perfusate summing up to about 70 and 100% of BD metabolized, which indicates the quantitative importance of the GSH dependent metabolism. In summary, the results demonstrate the relevance of an intrahepatic first-pass metabolism for metabolic intermediates of BD, which undergo further transformation immediately after their production in the liver before leaving this organ. Hitherto, the occurrence of this first-pass metabolism was only hypothesized. The findings will help to explain the drastic species difference between mice and rats in the carcinogenic potency of BD.