The tetrazolium salt method previously developed for the detection of xanthine oxidoreductase activity in unfixed cryostat sections has been validated for quantitative purposes. The specificity of the enzyme reaction was studied by incubating unfixed cryostat sections of rat liver in test medium containing the substrate hypoxanthine, in control medium that lacked the substrate, and in medium containing substrate and allopurinol, a specific inhibitor of xanthine oxidoreductase activity. The specific reaction rate was determined cytophotometrically by subtracting the amount of final reaction product generated in the control reaction from that formed in the test reaction. Highest specific enzyme activity in rat liver was found when the incubation medium contained 18% (w/v) polyvinyl alcohol, 100 mM phosphate buffer, pH 7.8, 0.45 mM 1-methoxyphenazine methosulfate, 5 mM tetranitro BT, and 0.5 mM hypoxanthine. Enzyme activity was present in liver parenchymal cells and in sinusoidal cells (endothelial and Kupffer cells) and was completely inhibited by allopurinol. A linear relationship was observed between the specific amount of final reaction product generated at 37 degrees C and incubation time at least up to 21 min, as well as section thickness up to 12 microns. Xanthine oxidoreductase activity, expressed as mumoles substrate converted per cm3 tissue/min, was 1.61 +/- 0.34 in pericentral areas and 1.24 +/- 0.16 in periportal areas. These values are similar to biochemical data reported in the literature. In conclusion, the tetrazolium method to detect xanthine oxidoreductase activity in unfixed cryostat sections of rat liver gives a reliable reflection of in situ activity.