We have reported previously that the apparent rate of peroxynitrite (ONOO(-) ) decay, as followed from its absorbance at 302 nm, decreases in the presence of hydrogen peroxide, mannitol and ethanol (Alvarez et al., 1995, Chem. Res. Toxicol. 8:859-864; Alvarez et al., 1998, Free Radic. Biol. Med. 24:1331-1337). Recently, two papers confirmed the observation and proposed that this slowing effect was due to the formation of absorbing peroxynitrate (O(2) NOO(-) ) as intermediate (Goldstein and Czapski, 1998, J. Am. Chem. Soc. 120:3458-3463; Hodges and Ingold, 1999, J. Am. Chem. Soc. 121:10695-10701). Peroxynitrate would be formed from the reaction of peroxynitrite-derived nitrogen dioxide with superoxide. Superoxide, in turn, would arise from the one-electron oxidation of hydrogen peroxide, or from the reaction of reductive radicals derived from mannitol and ethanol with dioxygen. In agreement with this concept, we show herein that under the conditions of our previous work, the slowing effect is prevented by superoxide dismutase and, in the case of mannitol and ethanol, by reducing the dioxygen concentration of the reaction solutions. Thus, superoxide formation is necessary for the decrease in the rate of absorbance decay. In addition, by simulations using known rate constants and absorption coefficients, we show that the slowing effect can be quantitatively accounted for by the formation of peroxynitrate.