The ventilatory response to an added external dead space was assessed in preterm babies, recovering from respiratory distress syndrome, immediately prior to extubation. All babies were ready for extubation as defined by routine clinical criteria. Baseline measurements of respiratory rate, tidal volume, and minute ventilation were made over a 2 min period using a computerized system consisting of a pneumotachometer connected directly to the proximal end of the endotracheal tube. The measurements were repeated after addition of an external dead space equivalent to 2 anatomical dead spaces (4.4 mL/kg body weight). Thirty-four babies were studied on 40 occasions. Twenty-four infants (60%) were successfully extubated and 16 (40%) required reintubation. Infants in the success and failure groups were matched for gestation at birth, postconceptional age and weight at the time of study, maximum ventilatory requirements, and treatment with methylxanthines. The added external dead space resulted in an increase in minute ventilation in 38 out of the 40 studies. Extubation success and failure groups were compared by expressing the minute ventilation after addition of the external dead space as a percentage of the baseline minute ventilation (%MV1). Successful extubation was associated with a higher median %MV1 compared with babies who failed extubation (156; range, 89.3 to 230; compared to 131; range, 75.2 to 165; P = 0.006). This test may be useful in deciding which babies could be successfully extubated.