The objective of this study was to compare circulating lipid docosahexaenoic acid [22:6(n-3), DHA] levels in term infants fed a powdered (CORN oil) or liquid (SOY oil) infant formula or human milk (HM). Infants whose mothers chose not to breast feed were randomly assigned to the CORN or SOY formula group. The formula fat differed in linolenic acid [18:3(n-3)] content: it was 0.8% for the CORN and 4.8% for the SOY. Linoleic acid [18:2(n-6)] was 31.5 and 34.2% fatty acids in the CORN and SOY formula, respectively. The formulas or HM were fed from birth through 8 wk of age, and growth and the plasma and red blood cell (RBC) phospholipid fatty acid composition was determined at 3 d, 4 wk, and 8 wk of age. Growth did not differ among groups. The plasma phospholipid and RBC phosphatidylethanolamine DHA was similar in the CORN and SOY formula groups at all ages. Plasma and RBC phosphatidylethanolamine levels of DHA were significantly lower in infants fed the CORN or SOY formula than in infants fed HM during wk 4 and 8. Plasma and RBC 22:5(n-6) was not increased in the formula groups at any age. The formula content of linolenic acid had no effect on the RBC or plasma DHA levels of the infants. The biologic or functional significance of the lower plasma and RBC DHA in infants fed formula rather than HM is unknown. The need for a dietary source of DHA and specificity of plasma or RBC phospholipid DHA as a measure of desaturation and elongation of linolenic acid in developing organs remains uncertain.