Compared to measurements, atmospheric models have overestimated [CIO]/[HCl] and [ClONO2]/[HCl] in the upper and middle stratosphere, respectively, and consequently have predicted lower [O-3] than observed. It is believed that a minor branch that produces HCl from the OH + ClO reaction can account for these discrepancies. Recent laboratory studies have indicated a 5 +/- 2% yield for this channel.(1) By performing box model sensitivity analysis calculations using the output from the LLNL-2D diurnally averaged stratospheric model, we quantitatively confirm that this reaction is the most prominent contributor to the model [ClO]/ [HCl] and [ClONO2]/[HCl] uncertainties in the upper stratosphere and that it is most effective in increasing [O-3] at higher latitudes during winter. Using theoretical methods, we examine the OH + ClO reaction mechanism, in which an initially energized HOOCl* complex is formed that can dissociate to HO2 + Cl (major) or HCl + O-2((1) Delta) (minor) products, redissociate to reactants, or be collisionally stabilized. Multichannel RRKM calculations guided by ab initio electronic structure calculations and the available kinetic data are presented. We show that the four-center transition state (TS3) for HCl production must lie at least 2 kcal/ mol below the reactants for the HCl yield to exceed 5%. Our ab initio relative energy of -2.3 +/- 3 kcal/mol for TS3 demonstrates that this minor HCl channel is mechanistically feasible. We also predict small pressure, temperature, and H/D isotopic dependencies for the minor channel yield and insignificant rates of complex stabilization under atmospheric conditions.