The role of drop collision-coalescence as a means of reducing drop number concentrations, and hence cloud condensation nucleus (CCN) concentrations as these are cycled in the stratocumulus-capped marine boundary layer, is investigated. We focus on the impact of this process on the mass-mean size of CCN in the absence of wet deposition, and compare this mechanism with size changes resulting from mass addition through aqueous chemistry. The modeling framework is a two-dimensional eddy-resolving model that includes explicit treatment of aerosol and drop spectra, as well as the solute transfer between drop size bins. The microphysical processes considered are droplet activation, condensation/evaporation, collision-coalescence, sedimentation and regeneration of particles following complete evaporation. It is shown that for a case exhibiting negligible wet deposition, collision-coalescence can significantly reduce drop concentrations (22% h-1) resulting in a measurable increase in particle mass-mean radius of about 7% h-1. In order to extend the validity of these results, trajectory analyses of parcel in-cloud residence times have been used together with box model calculations of collision-coalescence to explore the parameters affecting processing through collision-coalescence. This trajectory information is also used to deduce the extent of in-cloud conversion of SO2 to sulfate. Comparisons show that the two mechanisms may produce comparable rates of increase in the mean particle size under certain conditions. Results for remote marine conditions suggest that aqueous phase chemistry may have a greater impact at lower cloud liquid-water contents, whereas collision-coalescence may dominate at higher liquid-water contents, or for broader drop spectra. ¿ American Geophysical Union 1996