Recent studies have shown the relevance of air humidity to the provocation of bronchoconstriction by running. The present study was undertaken to ascertain whether the humid air breathed during swimming could explain the protective effect of swimming on the asthmatic. Nine asthmatic children 9--15 years old swam while inspiring dry (15--35% R.H.) or humid (80--90% R.H.) air administered in a random order, a week separating the two sessions. The exercise challenge was an 8-min tethered swim at a metabolic rate (VO2) of 29 ml.kg-1.min-1, minute ventilation (VE) of 34 L.min-1, and a heart rate (HR) of 161 beats.min-1. Ambient air and water temperature were 28 +/- 2 degrees C and 27 +/- 2 degrees C, respectively. Pulmonary functions were tested pre and post swimming. Exercise VE, VO2 and HR were similar under the two conditions. No reduction in any of the pulmonary functions (FVC,FEV1.0,MMEFR,MBC) was found after 5 and 10 minutes following the swimming exercise in either of the conditions. In contrast, a treadmill run of similar metabolic and ventilatory intensity induced bronchoconstriction when room air was dried to 25--30% R.H. It is suggested that, unlike running, swimming is of low asthmogenicity even when inspired air is dried to 25--30% at neutral temperatures.