A series of normal-valent compounds of potential interest with respect to atmospheric chlorine chemistry have been investigated using MCSCF ab initio methods with a good basis set. Comparisons of total energies of the compounds of interest with those of the radical precursors for complete active space wave functions including the bonding and antibonding orbitals of the molecules and the open shell orbitals of the radicals indicate that HOOCl, HOOOCl, and ClOOOCl are stable. Similar results for the compounds HOOH, ClOOCl, and HOOOH, which have been the subject of experimental or previous, more comprehensive, theoretical investigations, are presented for comparison. The predictions of stability for HOOCl and HOOOCl are of particular interest. HOOCl, which is the logical intermediate in the forward and backward reactions between HO + ClO and HOO + Cl, may be stabilized under some atmospheric and laboratory conditions. The role of HOOCl as an intermediate and the fact that the HOOCl adduct might be stabilized in certain parts of the stratosphere were suggested previously by Weissman et al. Also, Stimpfle et al. discussed HOOOCl as a possible intermediate in the HOO + ClO reaction but concluded on the basis of Benson's group additivity rules that its dissociation energy was probably too small for the molecule to be stable. The present results indicate that HOOOCl is sufficiently stable to be formed under some stratospheric conditions. At the MCSCF level of theory, HOOCl is predicted to have an electronic binding energy intermediate between those of ClOOCl and HOOH, while HOOOCl and ClOOOCl are predicted to have binding energies similar to that of ClOOCl.