1. Lewis and Dark Agouti (DA) rat strains (n = 4), models of human extensive and poor metabolizer phenotypes of debrisoquine/sparteine, respectively, were dosed with methoxyphenamine with and without prior administration of quinidine. Methoxyphenamine and its three metabolites, namely N-desmethylmethoxyphenamine, O-desmethylmethoxyphenamine and 5-hydroxymethoxyphenamine were quantified in 0-24 h urine. 2. The oxidative metabolic routes of methoxyphenamine which had been previously shown to involve the debrisoquine/sparteine isozyme, namely O-demethylation and aromatic 5-hydroxylation, were both significantly inhibited by quinidine in the two rat strains. 3. The oxidative metabolic route of methoxyphenamine which had been previously shown to not involve the debrisoquine/sparteine isozyme, namely N-demethylation, was not significantly inhibited by quinidine in either rat strain. 4. The Lewis strain pretreated with quinidine resembled the DA strain without such pretreatment in terms of O-desmethylmethoxyphenamine and 5-hydroxymethoxyphenamine in that the mean percentages of the dose excreted as these two metabolites and the mean O-desmethylmethoxyphenamine/methoxyphenamine and 5-hydroxymethoxyphenamine/methoxyphenamine ratios were similar to one another. 5. Ten days after quinidine administration to the Lewis strain of rat, all parameters of methoxyphenamine and its metabolites returned to normal. 6. A protocol involving substrate administration to Lewis strain rats with and without prior administration of quinidine could be developed as an attractive approach to screen substrates for metabolism in vivo by the debrisoquine/sparteine isozyme. Such an approach obviates interstrain differences.