There is an ongoing discussion in the chemical literature regarding the nature of the highly reactive hydroxyl radical formed from the reaction between ferrous iron and hydrogen peroxide (the Fenton reaction). However, the fundamental experiment of directly determining the source of the hydroxyl radicals formed in the reaction has not yet been carried out. In this study, we have used both hydrogen peroxide and water labeled with 17O, together with ESR spin trapping, to detect the hydroxyl radicals formed in the reaction. ESR experiments were run in phosphate buffer with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a spin trap, and either H2O2 or H2O labeled with 17O. The hydroxyl radical was generated by addition of Fe2+ ion to H2O2, or as a control, by photolysis of H2O2 in the ESR cavity. Observed ESR spectra were the sum of DMPO/.16OH and DMPO/.17OH radical adduct spectra. Within experimental accuracy, the percentage of 17O-labeled hydroxyl radical trapped by the DMPO was the same as in the original hydrogen peroxide, for either method of hydroxyl radical generation, indicating that the trapped hydroxyl radical was derived exclusively from hydrogen peroxide and that there was no exchange of oxygen atoms between H2O2 and solvent water. Likewise, the complementary reaction with ordinary H2O2 and 17O-labeled water also showed that none of the hydroxyl radical was derived from water. Our results do not preclude the ferryl intermediate, [Fe = O]2+ reacting with DMPO to form DMPO/.OH if the ferryl oxygen is derived from H2O2 rather than from a water ligand.