One hundred and four patients with ankylosing spondylitis were involved in a comparative trial carried out in one single centre. They were given either pirprofene (1000 mg/day), indomethacin (125 mg/day) or a placebo for 6 consecutive weeks. Response to treatment was assessed after 1, 2, 4 and 6 weeks on the basis of 4 subjective criteria and 6 standard measurements. The overall therapeutic results were judged by the investigator to be satisfactory in 87% of patients in the pirprofene group, 77% in the indomethacin group and 29% in the placebo group (p less than 0.01). A large number of patients in each group (including 34% on placebo) reported side effects. Treatment was discontinued on account of gastric pain in two patients (one on pirprofen, the other on indomethacin). No abnormalities were detected in laboratory tests.