In vitro aerosol exposure of epithelial cells grown at the air-liquid interface is an experimental methodology widely used in respiratory toxicology. The exposure depends to a large part on the physicochemical properties of individual aerosol constituents, as they determine the transfer kinetics from the aerosol into the cells. We characterized the transfer of 70 cigarette smoke constituents from the smoke into aqueous samples exposed in the Vitrocell® 24/48 aerosol exposure system. The amounts of these compounds in the applied smoke were determined by trapping whole smoke in N,N-Dimethylformamide and then compared with their amounts in smoke-exposed, phosphate-buffered saline, yielding compound specific delivery efficiencies. Delivery efficiencies of different smoke constituents differed by up to five orders of magnitude, which indicates that the composition of the applied smoke is not necessarily representative for the delivered smoke. Therefore, dose metrics for in vitro exposure experiments should, if possible, be based on delivered and not applied doses. A comparison to literature on in vivo smoke retention in the respiratory tract indicated that the same applies for smoke retention in the respiratory tract.