DHA and AA, which are components of breast milk but not infant formulas marketed in the United States and some other countries, are important components of the brain, and DHA is a major component of the retina. Also, many studies have demonstrated advantages of breastfeeding versus formula-feeding on subsequent cognitive and visual function; however, available data are insufficient to justify the conclusion that the presence of DHA and AA in breast milk is partially or soley responsible for the apparent advantages of breastfeeding. On the other hand, many studies of DHA (and AA)-supplemented versus unsupplemented formulas have shown clear advantages of the supplemented formulas on visual acuity at 2 and 4 months of age or neurodevelopmental status at 12 to 18 months of age. Although one logically may assume that these early effects may have long-term effects, this assumption is not warranted by the available data. One of the major problems is the difficulty of assessing visual and cognitive function of infants. Scores on standard neurodevelopmental tests at 1 year of age, for example, are only weakly correlated with performance at school age (when more definitive assessments are possible), and little is known about the predictability of later visual function from behavioral or electrophysiologic assessments of visual function early in life. Even prematurely born infants can synthesize DHA and AA and other omega-3 and omega-6 LC-PUFAs from the dietary EFAs, LA and ALA. Nonetheless, plasma, erythrocyte and brain lipid levels of DHA are lower in infants whose diets do not contain DHA. Whether more optimal intakes of ALA result in higher plasma and tissue levels of this FA is unclear. The breast-milk content of LC-PUFAs is not regulated by the mammary gland but, rather, reflects the concentrations of LC-PUFAs in maternal plasma lipids that, in turn, are dependent on maternal diet and, probably, maternal activities of the desaturases and elongases involved in converting dietary LA and ALA to LC-PUFAs. This occurrence suggests that some infants receive sufficient LC-PUFA to support normal rates of deposition, whereas others may not. Also, some infants probably can synthesize additional LC-PUFAs from the LA and ALA contents of human milk. Thus, depending on maternal diet and maternal and infant desaturase and elongase activities, some breastfed infants may receive less than adequate LC-PUFAs to support normal rates of deposition. Clearly, the role of LC-PUFAs in infant development is not a simple issue. Also, no foolproof method exists to ensure an adequate but not excessive intake. Thus, because some evidence shows that dietary LC-PUFA (DHA, AA, or both) as components of breast milk or formula confers at least transient developmental benefits, supplementation of infant formulas with LC-PUFAs is supportable provided that the supplements used are safe. The safety of all available supplements is unknown; however, some trials reveal few reasons for major concerns about the safety of single-cell oils, low-EPA fish oil, or egg-yolk phospholipid or triglyceride fractions.