Static and dynamic lung volumes, arterial blood gases, alveolar ventilation and ventilation-perfusion (v̇A/q̇) relationships were studied in 14 mild asthmatic patients and in 7 normal subjects (as controls) before and after fenoterol inhalation. Multiple nitrogen washout curves were analyzed by a bicompartmental distribution model, in order to assess the distribution of ventilation and v̇A/q̇ mismatch. At baseline, asthmatics showed mild airway obstruction and gas exchange impairment [forced expiratory volume in 1 s (FEV1) = 79% pred; PaO2 = 87.4; alveolar-arterial oxygen tension gradient (AaPO2) = 22.9 mm Hg]. By analysing nitrogen washout curves, an alveolar slow space representing 45.1% of total lung volume (vs. 36.8% in normals; p = 0.044) was identified; its alveolar ventilation per minute per unit lung volume (v̇A2/L2) was lower than in normals (p = 0.01). β-Agonist inhalation by the asthmatics, which reversed airway obstruction (FEV1 = 98% pred.; p < 0.001) and improved gas exchange (PaO2 = 92.6 mm Hg, p < 0.001; AaPO2 = 16.8 mm Hg, p = 0.003), led to a highly significant increase in v̇A2/L2 (p = 0.001). The improvement in PaO2 was associated with the increase in v̇A2/L2 (r2 = 0.39; p = 0.017), but not with the increase in FEV1. Lastly, the changes in FEV1 and v̇A2/L2 were not correlated with each other. We conclude that even in mild stable asthma there is substantial unevenness of ventilation, detectable by bicompartmental analysis of nitrogen washout curves, which is responsible for gas exchange impairment and is not related to common spirometric parameters. In addition, the improvement in gas exchange is probably due to the effect of fenoterol on the tributary airways of the alveolar slow compartment. This effect can be assessed by this simple method, which can be used in clinical pharmacology studies and in the follow-up of asthmatic patients.