A new technique with controlled interface generation allows separation and quantitation of enzyme inactivation by both solvent/aqueous interface and dissolved solvent. This has now been used in n-butanol, isopropylether, 2-octanone, n-hexane, n-butylbenzene, and n-tridecane. Ribonuclease was stable with all the solvent/aqueous interfaces studied. Chymotrypsin was mainly inactivated by the more hydrophobic solvent/aqueous interfaces, whereas lipase was only inactivated by the less hydrophobic solvent/aqueous interfaces. Urease was inactivated by some interfaces, but not all, without an obvious trend. Thus, the commonly expected simple relationship with solvent polarity (e.g., log P) does not apply when interfacial inactivation is determined specifically. Greater dissolved solvent inactivation occurred with the more polar solvents, though only a general trend was apparent with log P. A better correlation was noted with the Hildebrand solubility parameter. Interfacial effects are discussed with reference to enzyme molecular weight, denaturation temperature, hydrophobicity, and adiabatic compressibility.