In a randomised controlled trial 38 asthmatic children aged 2-11 yr who had not received regular oral or inhaled steroids during the previous year, were treated with a standard regime of nebulised salbutamol and intravenous aminophylline plus either hydrocortisone and oral prednisolone for 5 days, or placebo. The children were observed throughout their hospital stay and for 3 months afterwards. There was a greater fall in heart rates in the steroid treated group on the second day of treatment (mean diff. 16 beats/min) and at discharge (mean diff. 13 beats/min); p less than 0.025. Peak Expiratory Flow Rates recorded in 26 children, 13 in each group, showed more improvement on day 2 in those given steroids (mean diff 16% predicted); p less than 0.05. This difference was not apparent at discharge but 9 children treated with steroids were clinically wheeze-free when they left hospital compared with 3 in the placebo group, p less than 0.05. There were no differences in respiratory rate, pulsus paradoxus and arterial oxygen saturation. Trends in duration of hospital stay and relapse rate during the succeeding 3 months favoured active treatment. These findings support the use of systemic corticosteroids in addition to high dose bronchodilators to treat 'non steroid dependent' children hospitalised with acute severe asthma.