An investigation of ”interfacial resistances“ has shown that the liquid-liquid mass transfer rate in a stirred cell is reduced in the presence of both soluble and insoluble surface-active agents. For 200 r.p.m. stirring in each liquid phase the over-all coefficient, calculated on the basis of the water film, for the transfer of isopropyl alcohol from water to benzene was decreased from 6 × 10 −4 to 1·3 × 10 −4 cm/sec by the presence of a plasma albumin protein film of surface concentration 1·3 mg/m 2. This was the maximum reduction obtainable for all protein film concentrations and signifies the presence of an interfacial film strong enough to prevent ”surface renewal“. A ”surface clearing model“ is used to explain the effect of film strength (as measured by the surface modulus of compressibility) on surface renewal, and therefore on the mass transfer rate. Both film strength and rate of transfer were dependent on the type of aqueous substrate on which the protein film was spread. Substrates used with the benzene-water- isopropyl alcohol system were redistilled water, tap water, 0·01 N sodium hydroxide and 0·01 N hydrochloric acid. Similar results were obtained with the ethyl acetate-water system for the effect of plasma albumin films on the rate of solution of ethyl acetate in the water, using both redistilled water and tap water as substrates. For both of these systems measurements of the interfacial viscosity of the surface film indicated that this property had no direct effect on the transfer process. The effect of surface-active films on mass transfer rates has been further studied through a smaller number of experiments, including the spreading of an acid resistant protein (porcine mucosa pepsin) for acetic acid transferring from water to benzene, and also the spreading of films of cetyltrimethylammonium bromide and of octadecyltrimethylammonium chloride for the benzene-water- isopropyl alcohol system.