To examine the effect of corticosteroids on bronchial hyperresponsiveness, a randomised, double dummy, single blind crossover study was performed in 18 subjects with chronic asthma, comparing the effect of three weeks' treatment with inhaled beclomethasone dipropionate, 1200 micrograms daily, and oral prednisone 12.5 mg daily. The 12 week study began with a three week run in period of baseline treatment, which was continued unchanged throughout the study, and the two treatment periods were separated by a three week washout period. Patients kept daily Airflometer readings and attended the laboratory every three weeks for spirometry and a histamine inhalation test for determining the provocative dose of histamine causing a 20% fall in FEV1 (PD20). The mean FEV1 at the start was 1.9 litres (56% predicted). There was no significant change in PD20 with prednisone treatment, the mean PD20 being 0.56 and 0.59 mumol before and after treatment. There was, however, a significant improvement in PD20 with beclomethasone dipropionate treatment, the geometric mean PD20 being 0.38 and 1.01 mumol before and after treatment (p less than 0.001). There was a small but significant improvement in mean FEV1 after beclomethasone dipropionate treatment--from 1.9 to 2.2 litres--but no change after prednisone. Both medications produced significant and similar improvements in morning and evening Airflometer readings, post-bronchodilator improvement, and diurnal variation. Thus at doses that had similar beneficial effects on lung function beclomethasone dipropionate caused a significant improvement in bronchial hyperresponsiveness whereas prednisone caused no change. The superior topical anti-inflammatory effect of beclomethasone dipropionate may account for the different effects on bronchial hyperresponsiveness.