The results of electron-nuclear double resonance and electron paramagnetic resonance (EPR) studies on the hydrogen peroxide compound of yeast cytochrome c peroxidase are inconsistent with previous proposals for the source of the EPR signal in this compound, in particular with its identification with an aromatic amino acid radical such as would arise by oxidation of a tryptophanyl side chain. The present observations lead us to propose that the EPR signal is associated with a cluster containing at least one methionine and in which proximate side chains share the charge created by loss of one electron.