Enteral absorption by healthy male volunteers of cis-5,8,11,14,17-eicosapentaenoic acid (20:5(n-3), EPA) and cis-4,7,10,13,16,19-docosahexaenoic acid (22:6(n-3), DHA) from an ethyl ester enriched in EPA and DHA (K85) and from natural fish oil (TG30) was found to be similar after intake of equivalent doses. Thus, after administration twice daily for 14 days, the amounts of EPA and DHA in total serum lipids and in serum phospholipids were essentially identical for the two ester forms of the n-3 fatty acids. A dose-dependent relationship between intake and total phospholipid serum levels was observed for EPA, which increased 6.5-12-fold and 4.8-9-fold, respectively, but DHA reached a plateau at 2-2.5-fold over the control after supplementation. Arachidonic acid (AA) did not change significantly in total serum lipids, but fell 10-20% in serum phospholipids. The supplementation resulted in a 6.3-11-fold increase in the EPA/AA ratio. A significant and dose-dependent 25-45% reduction in linoleic acid (18:2(n-6), LA) was found in serum phospholipids, while a smaller decrease was observed in total serum lipids. 20:3(n-6), 20:3(n-9) and 22:5(n-6) also decreased significantly in serum phospholipids. Saturated fatty acids remained essentially unchanged. Some 35-47% decrease in serum triacylglycerols and approx. 10% decrease in cholesterol levels were also observed. In spite of the similar serum levels of EPA and DHA obtained in vivo, in vitro hydrolysis by porcine pancreatic lipase of K85 was 3-fold slower than hydrolysis of a glyceryl ester (TG85) similarly enriched in EPA and DHA, and 15-fold slower than the hydrolysis of the 3-fold less enriched TG30 with EPA and DHA predominantly in the 2-position. Under similar conditions release of AA from glyceryl ester and ethyl ester was essentially similar and approx. 1.5-fold faster than release of EPA and DHA from ethyl esters. In vitro hydrolysis of olive oil was 1.8-fold faster than hydrolysis of a natural fish oil (TG30) under similar conditions. Thus, the order of the rates of hydrolyses was olive oil > TG30 > TG85 > triarachidonin > LA ethyl ester > AA ethyl ester > ethyl ester K85. Our results show that in spite of differences in the rate of hydrolysis by lipase in vitro, the enteral absorption of EPA and DHA is as least as good from a synthetic ethyl ester highly enriched in EPA and DHA as it is from a natural triacylglycerol containing equivalent amounts of these fatty acids.