The purpose of the present study was to evaluate the effects of feeding human milk or milk formula on the fatty acid composition of plasma and red blood cell (RBC) lipids in at term small for gestational age infants (SGA) for the first 3 months of life. One group of infants received a formula with a linoleic:alpha-linolenic acid ratio of 10:1 (MF group). Another group served as control and received their own mother’s milk (HM group). Blood samples were taken at birth and at 1 week, 4 weeks, and 3 months of life. Plasma and RBC fatty acid composition were analyzed by gas liquid chromatography and results of total plasma lipids were expressed as concentrations by adding an internal standard. Concentrations of saturated and monounsaturated fatty acids increased in total plasma lipids with age in all infants. In contrast, those fatty acids decreased when results were expressed as percentages of total fatty acids. Long-chain polyunsaturated fatty acids (LCP) decreased regardless of how results were expressed, but the absolute concentrations of these fatty acids in plasma available for tissue accretion were greater than suggested by the percentage results. Plasma and RBC docosahexaenoic acid (22:6n-3) decreased in the MF group in comparison to the HM group. Arachidonic acid (20:4n-6)was lower in plasma of MF-fed infant but not in RBC phospholipids. We conclude that term SGA infants fed an adapted milk formula with a linoleic:alpha-linolenic acid ratio of 10:1 but devoid of LCP may lead to a low n-3 LCP status.