The purpose of the present study was to evaluate the responsiveness of normal human airways to beta-adrenoceptor stimulation using a forced oscillation technique. Seven normal volunteers (aged 25 +/- 2 year) were studied on three occasions, separated by weekly intervals, using a single-blind randomized design. On day 1, subjects were given cumulative doses of inhaled salbutamol (100, 200, 500, 1000 micrograms); and identical placebo was given on the other two visits. Respiratory oscillation impedance (Ros) was measured at baseline and 15 min after each dose increment. The coefficients of variation (CV) for short-term intra-individual variability on each placebo day were 7.5 and 9.5, and 9.6% for long-term variability (measured over all three visits). The 95% confidence values (2SD) for the change in Ros required to exclude natural variability were 0.39, 0.50 and 0.53 cmH2O l-1 s, respectively. There was a small fall in Ros in response to salbutamol although the mean maximum change (0.46 cmH2O l-1 s) was not significant (by ANOVA). Thus, the change in Ros (sensitivity) was no greater than the 95% confidence value for natural variability (reproducibility). Regression analysis also showed no evidence of a dose-response relationship for Ros.