The uptake and reaction of HNO3 with NaCl was studied at 298 K using a Knudsen cell coupled to a quadrupole electron impact mass spectrometer. Experiments were conducted using less than one layer of particles to clearly define the available reactive surface area. The uptake of HNO3 was observed to be faster initially than at longer reaction times. A new model is proposed that incorporates reactions at two different types of sites: (1) steps and edges on the NaCl surface holding surface adsorbed water (SAW) and (2) dry terrace sites. The initial, more rapid uptake is attributed to reaction with both types of sites on a fresh NaCl surface, while the slower uptake at longer reaction times is due only to reaction at the steps and edges where new reaction sites are generated by the SAW-assisted recrystallization of the NaNO3 product. The initial uptake coefficient based on data at short reaction times is gamma(0) = (2.3 +/- 1.9) X 10(-3), in agreement within the experimental errors with a value of gamma(0) = (1.1 +/- 0.4) x 10(-3) (2s) derived from the dependence of the uptake coefficient on the initial HNO3 concentration. The fraction of the surface area covered by sites holding SAW was estimated to be approximately 50% based on the uptake coefficient, averaged over three cycles of exposure to HNO3, gamma = (1.0 +/- 0.8) x 10(-3) (2s). The data suggest that the reaction of HNO3 with effluoresced sea salt particles is less important than previously thought relative to reactions with N2O5 and ClONO2, which generate chlorine atom precursors.