The field measurement of dry deposition still represents a difficult task. In our approach, a I to 2 cm thick layer of water in a petri dish with a diameter of 22 cm, serves as a surrogate surface. The atmospheric constituents taken up by the water can be analyzed chemically by the same procedure as for the wet deposition samples. In contrast to solid surrogate surfaces, water exhibits the following advantageous properties: low and constant surface resistance, high sticking coefficient for aerosols, and predictable sorption behavior for gases. Consequently, the deposition rates measured to the wet surface are generally higher, by up to a factor of 4 for NH 4 •,CJ-, NO,and S0,2-, than those to a dry surface, but still smaller than the concurrent wet deposition rates. We observed the following average dry deposition rates in µmo! m-2 ct-•: NH 4 • 48.3, Ca2+ 40.7, Na• 15.8, Mg 2 • 8.4, K• 4.2, H-Aci 36.4; S0,2-57.2, CJ-39.2, NO,-34.5, HSO,-5.7, formate 4.0; acid soluble metals: Fe 2.8, Zn 0.60, Cu 0.11, Pb 0.073, Cd 0.0022. The soluble fraction of Zn, Cd, Cu, Pb and Fe in the dry deposition varied with the pH of the water phase corresponding to the adsorption tendency of these metals to oxide surfaces. The sampling method also allows tracing of regionally and locally emitted atmospheric pollutants. In our study the local pollution sources included road salting, construction work and a refuse incinerator. Finally, chemical reactions occurring in the atmosphere, such as the conversion of c1to HCI by HN0 3 or the oxidation of SO,, can be identified by evaluating the data. The method proposed is relevant to measure reproducibly the dry deposition of a variety of compounds to water bodies and moist vegetation.