A total of 35 iron balances were performed using a milk formula containing 6.25 +/- 1.23 mcg./dl. of iron. Twenty balances were done during the first 3 days of life (group 1) in low birth weight infants with 36.2 +/- 2.6 weeks of gestation and with birth weights of 2.13 +/- 0.28 kg. Thirteen of these infants received a calcium lactate supplement of 800 mg./kg./day. Ten balances were carried out at 13.9 +/- 4.12 days of life (group II) in LBWI with 36.0 +/- 2.1 weeks of gestation and with birth weights of 2.08 +/- 0.22 kg. Five balances were performed in a heterogeneous group (group III) of patients with iron-deficiency anemia (transferrin saturation: 6.3 +/- 2.9%). In group 1, the absorption coefficient for iron was 48.5 +/- 13.6% and of the 562. +/- 155.9 mcg./kg./day of iron ingested, 577.7 +/- 277.7 mcg./kg./day were retained. There were no statistically significant differences between groups I and II, but there were significant differences for ingestion, excretion and iron retention (p less than 0.001). In group I, iron absorption was not modified by the addition of calcium lactate. There was a positive significant correlation (p less than 0.001) between ingestion and net iron retention in groups I and II. Balances performed on five anemic children showed absorption coefficients of 10.5, 11.5, 41.6, 60.0 and 74.2% respectively. In view of our data, recommendations as to the optimal iron supplementation of cow's milk and as to the most desirable time to initiate it are analyzed.