Using RIA technique, response patterns of ACTH-adrenocortical system to short term large dose corticosteroid therapy for neurosurgical patients were investigated. The subjects were consisted of 20 males and 13 females, ageing ranging from 10 to 60 years old. The subjects were divided into three groups according to mode of steroid administration. In the first group, 12 mg/day of betamethasone were administered for 5 days and cancelled without gradual reduction. In the second group, 12 mg/day of betamethasone were administered for 4 days and 6 mg/day of betamethasone were given for following 2 days. In the third group, 12 mg/day of betamethasone were administered for 3 days and gradual reduction of betamethasone was given for following 4-5 days. Response of plasma cortisol level to synthetic 18-ACTH were measured in all patients for 3 days after betamethasone administration cancelled. The results led to the following conclusion. 1) The second and third groups had not any clinical symptom after steroid cancelled, but one half cases of the first group had mild headache, nausea for 2 days after cancelled. 2) Recovery process of adrenocortical function in the second group is most rapidly, compared with that in other two groups. 3) In the postoperative steroid therapy in neurosurgery, it could be concluded that 4 days massive administration of steroid followed by stepwise reduction in 2 days would be functionally well tolerated by adrenocortical system.