To quantitate the formation of hydroxyl radicals (HO.) in ischemia and reoxygenation, dimethyl sulfoxide (DMSO) was added to "trap" evolving HO. in normal, in ischemic, and in ischemic and reoxygenated rat kidney slices, incubated in short-term organ culture in vitro. Hydroxyl radical generation was measured as the accumulation of the specific product of DMSO oxidation by HO., methane sulfinic acid (MSA) in the kidney tissue and surrounding medium using a new colorimetric assay. A mean difference of 7 nmol cumulative HO./gram tissue was detected in rat kidney slices subjected to ischemia and reoxygenation. This amount of HO. generation was not significantly greater than that found in nonischemic or in ischemic but not reoxygenated control tissues, and does not appear to represent the highly toxic burst of HO. radicals implied in current theoretical discussions of reperfusion injury. However, the addition of EDTA chelated iron (1:1) to the incubation medium led to marked postischemic HO. generation. We conclude that clearly toxic numbers of HO. radicals are not formed during reoxygenation in rat kidney slices, either because there is insufficient iron, because only a small fraction of cells in the kidney tissue make oxygen radicals, or because cellular defenses against HO. formation are more powerful than currently appreciated.