Forty-one patients with periarthritis of the shoulder were entered into a double-blind, parallel comparison of naproxen and indomethacin. Both drugs were given twice daily, naproxen 250 mg in the morning and 500 mg at night and indomethacin 50 mg b.d. Twenty patients received naproxen and 21 received indomethacin. Patients were treated for four weeks. Both drugs significantly reduced pain and patients subjective impression of shoulder restriction, but had little significant effect on objective measurements of mobility. Fourteen patients from the naproxen group and 16 patients from the indomethacin group reported side-effects, the most common being nausea and headache. Three patients stopped naproxen and five patients stopped indomethacin because of side-effects. In conclusion, both drugs were equally effective in treating the pain of periarthritis of the shoulder but did little to change the partial loss of movement associated with the disorder.