For 11 drugs it was investigated whether tissue distribution in vivo can be predicted by use of binding data obtained in vitro. The selection of drugs represented a broad spectrum of physicochemical and pharmacokinetic properties thought to be important for distribution of drugs in vivo. The extent of binding to plasma and to tissue-homogenates of rabbits was determined in vitro. The drug concentrations in plasma, liver, lungs, kidneys, and skeletal muscle of rabbits were determined in vivo after i.v. administration of the drug. The tissue-plasma partition ratios measured in vivo were compared with the theoretical tissue-plasma partition ratios calculated from the in vitro binding data. For all drugs investigated the muscle-plasma partition ratio could be reasonably well predicted by the in vitro binding data. In liver, lungs, and kidneys good agreement was found between measured and predicted tissue-plasma ratio for anionic drugs; marked differences, however, were observed between measured and predicted tissue-plasma ratios of lipophilic cationic drugs. A significant correlation was found between binding of drugs to muscle tissue in vitro and the volume of distribution of the unbound drug (Vf), opening the possibility to approximate Vf from in vitro binding studies with rabbit muscle tissue.