The enzymatic transesterification of palm olein was conducted in a low-moisture medium with nonspecific and 1,3-specific lipases from microbial sources. The enzymes were first immobilized on Celite, ]yophilized for 4 h and then added to a reaction medium that consisted of 10% (wt/vo[) palm olein in water-saturated hexane. The catalytic performance of the enzymes was evaluated by determining the changes in triglyceride (TG) composition and concentrations by reverse-phase highperformance liquid chromatography (HPLC) and the formation of free fatty acids by titration. Studies with lipase from Candida rugosa showed that the degree of hydrolysis was reduced by drying the immobilized preparation and that the best drying time was 4 h. In all cases, the transesterification process resulted in the formation of PPP, a TG initially undetected in the oil, and increases in the concentrations of OOO (1.3-2.1-fold), OOL (1.7-4.5-fold), and OLL (1.7-4.3-fold), where P, O, and L are palmitic, oleic, and linoleic acids, respectively. SOS (where S is stearic acid), another TG not detected in the oil, was synthesized by Rhizomucor miehei and Pseudomonas lipases, with the latter producing more of this TG. There was a corresponding decrease in the concentrations of POP, PLP, POO, and POL. PPP concentration ranged from 1.9% (w/w) for Mucorjavanicus [ipase to 6.2% (w/w) for Pseudomonas lipase after 24 h. The greatest degree and fastest rate of change were caused by Pseudomonas lipase, followed by the enzymes from R. miehei and Aspergillus niger. The effects of transesterification and hydrolysis of palm olein by the various lipases resulted in changes in the overall degree of saturation of the triglyceride components. There seems to be no clear correlation between the enzyme positional specificity and the products formed. Possible mechanisms for the formation of PPP, OOL, OLL, OOO, and SOS are discussed. JAOCS 72, 633-639 (1995).