The measurement of urinary mutagenicity is a non-invasive monitoring tool which often reflects an animal's recent exposure to genotoxic agents. Although studies in man are indispensable for monitoring industrial and/or environmental exposure to genotoxins, a sensitive laboratory animal model is necessary for mechanistic studies on the role of specific chemical exposure in altering urinary mutagenicity. The objective of this study was to enhance the sensitivity of the methodology used for detecting urinary mutagenicity in rats by using XAD-2 resin to extract and concentrate the urine and a microsuspension-modified Ames test to quantify mutagenicity. The polycyclic aromatic hydrocarbon benzo[a]pyrene (BP) and the aromatic amine 2-acetylaminofluorene (AAF) were used as test compounds. Under the conditions of our study, AAF administered to rats by gavage at doses of 1 mg/kg or higher induced a dose-dependent increase in urine mutagenicity. The greatest mutagenic response was seen when S9 was present during the microsuspension-modified Ames test and beta-glucuronidase (BG) was not included. Similarly, BP administered to rats by gavage at doses of 10 mg/kg or higher induced a dose-dependent increase in urinary mutagenicity. The relative importance of BG and S9 were quite different with BP than with AAF. With BP, mutagenicity was greatest when both S9 and BG were present during the microsuspension-modified Ames test, and least with S9 and without BG. In both AAF- and BP-treated animals, extraction of the urine on XAD-2 resin markedly enhanced the mutagenic response compared to neat urine, but partitioning of the XAD-2 eluate into methylene chloride always diminished the mutagenicity of the urine extract. The results demonstrate the sensitivity and reproducibility of rodent urinary mutagenicity assays when XAD-2 resin is used to extract and concentrate the urine and a microsuspension-modified Ames test is used to quantify mutagenicity. This sensitive method should facilitate mechanistic studies on the roles of specific environmental agents in affecting urinary mutagenicity and, in addition, may be used during acute, subchronic and chronic rodent bioassays as a non-invasive in vivo indicator of genotoxic exposure.