Respiratory distress syndrome of the preterm baby is believed to be caused by a deficiency of pulmonary surfactant and leakage of plasma into the alveolar spaces. Since the two pathogenetic factors seem to be inter-related, we postulated that peroxidation of surfactant by plasma iron could be the linking mechanism. We obtained cord blood samples from 22 preterm babies (mean gestational age 32.2 [SD 2.7] weeks) and 24 term babies (40.1 [1.6] weeks), and venous blood samples from 18 healthy adults. No adult had detectable non-protein-bound iron in the plasma, but 10/21 (48%) preterm babies and 6/24 (25%) term babies had detectable concentrations (rate difference 23% [95% Cl -5 to 51%], p = 0.20). Transferrin and haptoglobin concentrations were higher and free haemoglobin concentrations lower in adults than in babies (p < 0.005). Only transferrin differed significantly between term and preterm babies. Plasma from all 18 adults and from 23 (96%) term babies inhibited iron-catalysed lipid peroxidation of pulmonary surfactant liposomes. By contrast, plasma from 11 (50%) preterm babies stimulated such peroxidation (difference in stimulation rate 46% [20-71%], p < 0.005 for preterm vs term babies); the ability to stimulate peroxidation was related to the presence of non-protein-bound iron (p < 0.001). Peroxidation decreased in the babies when apotransferrin was added to plasma and in all subjects when alpha-tocopherol was incorporated into the surfactant liposomes. Lipid peroxidation of surfactant may contribute to the pathogenesis of respiratory distress syndrome. Possible therapeutic approaches are increasing babies' iron-binding capacity by plasma transfusions and increasing the antioxidant capacity of commercial surfactant.