A survey of 118 patients seen in the last twenty years in Newcastle upon Tyne forms the basis of this report. All of these 118 patients fulfilled clearly defined clinical, electrophysiological and pathological criteria for the diagnosis of polymyositis: muscle pain, weakness and characteristic EMG and/or muscle biopsy 55%; and characteristic muscle biopsy 17%; muscle weakness and characteristic EMG 7%; muscle weakness and pain, and raised serum CK activity in an established collagen-vascular disease 5%. A smaller group of 25 patients were selected in whom the clinical characteristics, EMG, muscle biopsy and serum enzyme levels were all completely diagnostic of polymyositis. The patients were followed for two months to twenty-six years, with a mean follow-up duration of six years. Analysis was made of the features at presentation and during the course of the illness, and of prognostic factors bearing upon the disability, response to treatment and mortality. Cases were classified according to the system of Rose and Walton (1966). Groups I, II, and III each constituted approximately one-third of the total cases, while only 8% of all cases were associated with carcinoma. The female to male ratio was 1.4:1. Though cases were seen in all age groups, the largest number was in the sixth decade. The sedimentation rate was raised in 55% of cases. Electromyography was characteristic of polymyositis in 45% of cases, and in only 11% was it normal. The serum creatine kinase activity was raised in 64% of cases. There was no correlation between the extent of these abnormalities and the degree of weakness or disability. 65% of muscle biopsies had changes with inflammatory infiltration virtually diagnostic of polymyositis. 17% of cases had a normal muscle biopsy. Most of the patients (89%) were treated with high-dose prednisone therapy, commencing with 30-100 mg/day, gradually reducing to a maintenance dose of 5-15 mg/day over two or three months. All clinical groups showed considerable improvement in average disability with time on "high dose" corticosteroid therapy, the maximum improvement occurring within the first three years. The degree of improvement in disability was considerably less in those inadequately treated, though the mortality rate was similar in the two groups. 66% of all survivors had essentially no functional disability at follow-up three or more years later, and in the majority of these cases the disease appeared to have burned itself out. 33% of cases had significant disability after three years, and in half of these the disease appeared to be still active.