Response to treatment was evaluated prospectively in 58 patients with Graves’ ophthalmopathy. Prednisone, administered to 44 patients, resulted in a favourable response in 29 patients (66%): excellent in 3, good in 10 and fair in 16 subjects; 15 patients (34%) did not respond. Orbital irradiation was administered to 39 patients of whom all except 4 had been treated previously with steroids. Six months later, 25 patients (64%) had responded favourably (excellent in 0, good in 9 and fair in 16 subjects) and 14 patients had not responded (36%). A favourable response to prednisone or irradiation was reflected in a general shift to lower grades in each of the classes 2–6 of the NOSPECS system. Responders were not different from non-responders with regard to age, sex, NOSPECS classification, ophthalmopathy index, interval between the onset of eye disease (or of thyroid disease if present) and the start of treatment, or the presence of thyroid disease. In conclusion, (1) the ratio of responders to non-responders is 2:1 for both prednisone treatment and orbital irradiation in Graves’ ophthalmopathy, (2) the response is seen in all classes of the NOSPECS system, and (3) the response is not dependent on age, sex, severity or duration of eye disease, or the presence of thyroid disease.