The effect of the nature of the filling gas on the persistence of microbubbles in the bloodstream was studied. All the microbubbles were covered with the same shells. Various perfluorocarbons and perfluoropolyethers alone and as mixtures with nitrogen were used as the filling gases. The persistence time of microbubbles in the bloodstream tau increased with the molecular weight of the filling gas, from approximately 2 min for perfluorethane, to > 40 min for perfluorodiglyme, C6F14O3, and then decreased again to 8 min for C6F14O5. An acceptable ultrasound scattering efficacy was exhibited by the filling gases with intermediate molecular weights that possessed both a high saturated vapor pressure and a comparatively low water solubility (Ostwald coefficient). On the basis of the experimental data, it is concluded that the microbubble persistence tau is controlled primarily by the dissolution of microbubbles and not by the removal of the microbubbles by the reticular endothelial system. Although the qualitative experimental trends are in good agreement with the theoretical model developed previously, there are some quantitative differences. Possible reasons for these differences are discussed.