We measured vertical gradients of nitric acid vapor and nitrate aerosol near the Boulder Atmospheric Observatory (BAO) tower, 20 km north of Dever, Colorado. This site was usually downwind of two large ammonia sources, a cattle feedlot and a chicken farm . The gradients we observed were very different from measurements at other sites: the aerosol gradient was in many cases steeper than that of the vapor, even though both experience and theory predict that aerosol dry deposition should be much slower (with a correspondingly flatter gradient) than that for nitric acid vapor. In some cases we even saw an apparent emission of nitric acid vapor, with the highest concentrations near the surface. These results are consistent with the predictions of a model which couples the evaporation of ammonium nitrate aerosol to the rapid dry deposition of nitric acid vapor. As deposition depletes the vapor near the surface, evaporation of the aerosol resupplies it. This produces a steep arerosol concentration gradient, even though the aerosol itself is not the form which is transported to the the receiving surfaces. One result of this coupling is that the traditional application of deposition velocities and measured concentrations to estimate dry nitric acid deposition may cause significant errors, since the individual species fluxes are not conserved (and may even have different signs under some conditions). The total nitrate flux is conserved, however, and may be the most predictable of the fluxes. ¿ American Geophysical Union 1988