One hundred and two (92 per cent) of 111 immature patients in whom idiopathic scoliosis had been treated with a Milwaukee brace were followed to determine the effectiveness of the brace in preventing progression of the scoliosis. The average time from cessation of bracing until the latest radiographs were made for the patients who were managed non-operatively was six years and four months. The average progression of the curve, from the time of initial bracing until use of the brace was stopped, in the eighty-eight patients who were included in the statistical analysis was 4 degrees. The curve continued to progress an average of 5 degrees after use of the brace was stopped in the patients who did not have an arthrodesis. Forty-two patients (48 per cent) had more than 5 degrees of progression at the time that use of the brace was stopped. Thirty-seven patients (42 per cent) had an operation or a curve of sufficient magnitude to warrant operative intervention.The maximum correction of the Cobb angle in the brace had prognostic importance for progression of the curve. The patients in whom the curve did not progress or who did not need operative intervention had had an average correction of 20 per cent, while the patients who had a failure had had an average correction of 8 per cent. The patients who eventually had the indications for an arthrodesis were, on the average, one year younger (eleven years and nine months) and had a curve of a larger magnitude at the time of bracing than the patients who did not need an arthrodesis.The findings of this study do not agree with previously reported favorable results with bracing and raise questions about whether the natural history of progressive idiopathic scoliosis is truly altered by use of the Milwaukee brace.