It has recently been shown that the omega 3 fatty acid status in humans can be predicted by the concentration of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids in plasma phospholipids [Bjerve, K.S., Brubakk, A.M., Fougner, K.J., Johnsen, H., Midjthell, K., and Vik, T. (1993) Am. J. Clin. Nutr., in press]. In countries with low intake of omega 3 fatty acids, the level of EPA in plasma phospholipids is often only about one-fifth the concentration of DHA. The purpose of this study was to investigate whether this difference in the concentration of these two fatty acids was due to a selective loss of EPA relative to DHA or to a lower dietary intake of EPA. Seven female volunteers ingested four grams of MaxEPA daily for 2 wk and in the following 4 wk they ate a diet almost completely devoid of the long-chain omega 3 fatty acids. The concentrations of the omega 3 fatty acids in the plasma cholesteryl esters, triglycerides and phospholipids and the high density lipoprotein phospholipids were examined at weekly intervals throughout the study. There was a more rapid rise in the concentration of EPA than in DHA levels in the supplementation period in all lipid fractions, but there was a disproportionate rise in DHA relative to EPA in the plasma lipids compared with the ratio in the supplement. In the depletion phase there was a rapid disappearance of EPA from all fractions, such that pre-trial levels were reached by one week post-supplementation.(ABSTRACT TRUNCATED AT 250 WORDS)