We investigated the relationship between airway calibre and the dose and distribution of inhaled aerosol in ten normal and six asthmatic subjects. Subjects inhaled saline aerosol containing 99mTcO4 delivered from a nebulizer connected to a dosimeter, and the lung fields were scanned with a gamma-camera. Right lung dose (RLD) was calculated as percentage of total dose. Intrapulmonary distribution was measured as penetration index (PI) (peripheral zone counts/central zone counts). Asthmatics had a significantly lower PI than normal subjects and there was a linear relationship between PI and baseline specific airway conductance (sGaw, p less than 0.001), and forced expiratory volume in one second (FEV1, p less than 0.05). After bronchodilatation with salbutamol (delta sGaw 101 +/- 31%, mean +/- SEM), PI increased from 0.73 +/- 0.11 to 1.09 +/- 0.15 (p less than 0.05); after bronchoconstriction with methacholine (delta sGaw 62.6 +/- 2.9%), PI decreased from 1.42 +/- 0.24 to 1.06 +/- 0.22 (p less than 0.05). Changes of PI were correlated with changes in sGaw and FEV1 (n = 20, p less than 0.001) but changes of RLD and changes in airway calibre were not. The distribution of inhaled aerosol, but not the dose, is largely dependent on airway calibre. The differences in PI between normal and asthmatic subjects may at best be explained by the differences in central airway calibre.