The radioaerosol method was used to measure the effect of saline aerosol on the rate of clearance of secretions from the lung. Two trial runs were done in each of 7 patients with chronic obstructive lung disease (mean +/- SD ratio of 1-sec forced expiratory volume for forced vital capacity, 50 +/- 15 per cent), a control run and a saline run. In both runs, 5.0 +/- 0.7-micron polystyrene particles tagged with 99mTc were inhaled under controlled conditions by the patients, and their subsequent clearance was monitored for 6 hours by whole lung counters. Scanning was also done with a gamma rectilinear scanner. The saline run was identical to the control run except that 30 min after inhaling the radioaerosol the patient inhaled an aerosol of hypertonic (1.21 M) saline for 11 min from an ultrasonic nebulizer. Although the initial distribution of the radioaerosol along the airways was the same in both runs, whole lung clearance during the first 50 min was twice as fast after the inhalation oft he hypertonic saline aerosol as in the control (P less than 0.01). The mean weight of sputum produced was higher (P less than 0.05) in the saline run. The number of coughs in the 2 runs was the same.