My colleagues and I compared the biochemical status and rates of growth of three groups of preterm infants: one group was fed milk obtained early from mothers of preterm infants; one group received milk produced during the mature stage of lactation by mothers of term infants; and one group received a whey-based infant formula. Sixty healthy preterm infants with birth weights of 1600 g or less were randomly assigned to one of the three feedings groups. The 20 infants in each group were followed until they reached a weight of 1800 g. The mean (+/- S.E.M.) number of days required to regain birth weight was similar for infants receiving the formula (10.3 +/- 0.8) and those receiving milk from mothers of preterm infants (11.4 +/- 0.8); both were significantly less than the number (18.8 +/- 1.7) for infants receiving milk from mothers of term infants (P less than 0.001). Subsequent rates of weight gain were greater for the groups receiving formula (27.0 +/- 0.8 g per day) and milk from mothers of preterm infants (23.7 +/- 1.1) than for the group receiving milk from mothers of term infants (15.8 +/- 0.8) (P less than 0.001). Similarly, the average increments in crown-to-heel length and in the head circumference were significantly greater for the groups given formula and milk from mothers of preterm infants (P less than 0.005 and P less than 0.001, respectively). These data indicate that feeding with either milk from mothers of preterm infants or a whey-based infant formula results in more appropriate growth in preterm infants than feeding with milk from mothers of term infants.