At neutral pH, compound I of Arthromyces ramosus peroxidase (ARP) was stable and was reduced to ferric ARP without apparent formation of compound II upon titration with ascorbate or hydroquinone. In the titration experiments, compound II was seen as an intermediate only at alkaline pH. However, measuring a difference spectrum in the Soret region by a stopped-flow method, we found that compound II was formed during the catalytic oxidation of ascorbate even at neutral pH. Using an EPR spectrometer with a microflow system, we measured the steady-state concentration of benzosemiquinone formed in the ARP-catalyzed oxidation of hydroquinone. The results clearly showed that ARP catalyzes the oxidation of hydroquinone by a one-electron-transfer mechanism, as does horseradish peroxidase. These observations led to the conclusion that compound I is reduced to compound II through a one-electron reduction by ascorbate or hydroquinone. Therefore, we concluded that ARP compound II is unusually unstable and is rapidly reduced to ferric enzyme without accumulation in the titration experiment. The unusual instability of ARP compound II is explained in terms of the high reduction potential of compound II. The reduction potentials (E0') of compounds I and II were measured at several pH values from redox equilibria with potassium hexachloroiridate on the basis of E0' = 0.90 V for the IrCL6(2)-IrCl6(3)- couple. These values were determined to be 0.915 and 0.982 V at pH 7, respectively, and decreased with increasing pH. This pH dependence was markedly changed by the buffer concentration.(ABSTRACT TRUNCATED AT 250 WORDS)