The efficiency of different preparations of lipases was evaluated in organic solvents. Lipases from Humicola lanuginosa, Candida rugosa, Rhizomucor miehei and Pseudomonas cepacia were adsorbed onto the surfactant sorbitan monostearate (Span 60) and the specific activities were compared in hexane to crude powder (used straight from the bottle) and lipase freeze-dried from buffer solution. Lipases adsorbed on the surfactant were "activated" 1.9- to 150-fold compared to the crude lipase. The solubility of the lipase-surfactant preparation in the reaction media was extremely low and the preparation contained aggregates of micrometer size. In further comparisons lipase from H. lanuginosa was freeze-dried in the presence of KCl, crown ethers, immobilised by entrapment into a sol-gel and immobilised on porous polypropylene support (Accurel EP-100). Addition of potassium chloride before freeze-drying of the lipase increased the activity up to 46-fold compared to crude powder. The additive probably worked as an immobilisation matrix for the lipase. When 18-crown-6 was added to the lipase before freeze-drying a 40-fold increase in activity was achieved. In this case a low amount of additive (0.4 mg crown ether/g protein) was needed for activation indicating that specific interactions were involved in the activation. In order to obtain maximal activity, immobilisation on Accurel EP-100 and entrapment into a sol-gel were the best methods to use. The activities were 400- and 320-fold better than that of crude powder. The high activities obtained were due to an improved dispersion of the catalyst in the organic media. The protein loading that could be used when the lipase was adsorbed onto Accurel EP-100 was much higher than what could be used when the lipase was entrapped into a sol-gel. This makes the adsorption technique the best for practical applications.