Adrenocortical function was assessed by the intravenous short synacthen test in 22 control subjects and 68 patients admitted to hospital with acute severe asthma. The cortisol increment was subnormal in 19 of the 68 asthmatics. This included 11 of the 14 patients on continuous oral steroids, seven of the 29 patients who had had occasional courses of oral steroids, one of the seven on inhaled steroids only, and none of the 18 who had had no steroids. Adrenal suppression was greatest in those patients taking oral steroids in divided daily doses. Nineteen of 43 patients were on or had taken oral steroids in this fashion. Of those 19 patients with low cortisol increments only one half had received supplementary steroids in the 24 hours preceding admission. Based on the synacthen test, serum DHEA-SO4 values were not a good discriminant of adrenocortical function. Adrenal insufficiency may be an important cause of death in acute severe asthma in New Zealand.