To test the hypothesis that premature infants in whom extubation fails in the first 10 days of life have low volume lungs, functional residual capacity (FRC) was measured in the first hour after extubation. Once extubated, infants received the appropriate level of inspired oxygen necessary to maintain acceptable arterial oxygen saturation. After humidification, oxygen was bled into a headbox, and FRC was assessed using a helium gas dilution technique and a specially designed infant circuit. The results were related to extubation failure, which was diagnosed when the infant required nasal continuous positive airway pressure or re-intubation and ventilation within 48 hours. The latter two forms of respiratory support were instituted by the clinical team, whenever the infant developed recurrent or severe apnea or respiratory acidosis. Infants were eligible for entry into the study when born prematurely and extubated within the first 10 days of life. Twenty infants initially ventilated for respiratory distress syndrome at a median gestational age of 29 weeks (range, 26-36 weeks) were studied at a median postnatal age of 3 days (range, 1-7 days). All were receiving theophylline. Extubation failed in seven infants, who did not differ significantly from the rest of the cohort regarding gestational age, birthweight, postnatal age, or inspired oxygen concentration (F(I)O2) at extubation, but their maximum F(I)O2 during ventilation was higher than in those infants who did not require reintubation (P < 0.05). In the infants who failed extubation, the median FRC was 19 ml/kg (range, 12-27 ml/kg), which was lower than that of the infants in whom extubation was successfully accomplished (median, 28 ml/kg; range, 19-37 ml/kg; P < 0.01). An FRC of less than 26 ml/kg had a sensitivity of 71% and specificity of 77% in predicting extubation failure. These results support the hypothesis that a very low lung volume relates to extubation failure in the first 10 days of life.