Two hundred and twenty-two tumors in 218 patients were collected from different institutions and doctors. Eight patients had died less than eighteen months after the diagnosis was made, one of an unrelated disease, the others as a result of their tumor. The remaining 210 patients with 214 tumors were followed for an average of 9.9 years, the range being two to thirty-four years. In each case, skeletal maturity, location of the lesion, roentgenographic findings, histological characteristics, and treatment were analyzed. Skeletal maturity, as evidenced by closure of the epiphyses, was present in all patients. This observation may explain the predominance of females (27:11) in the patients less than twenty years old. More than half (55 per cent) of the lesions were located in the lower end of the femur, the upper part of the tibia, and the lower end of the radius. Other sites were the sacrum, pelvis, proximal end of the femur and fibula, and the small bones of the hands and feet. The roentgenographic findings were sufficiently characteristic to be helpful but not diagnostic. They were influenced by previous treatment of fracture. The histological features permitted grading as suggested by Jaffe, Lichtenstein, and Portis30. However, the grading was not of prognostic value although it did serve to alert the surgeon to the possibility that the tumor was malignant. A consistently accurate prediction as to local recurrence or pulmonary metastasis could not be made on the basis of the histological grade. The treatment of the tumors in this series was performed by many surgeons and varied considerably. All patients had a preliminary biopsy. The primary surgical treatment ranged from curettage or excision, with or without bone-grafting, to amputation. The success of surgical treatment, whether by curettage or resection, appeared to depend on the completeness with which the tumor was removed. After primary curettage, thirty-five of forty-five tumors recurred; whereas, after primary curettage combined with bone-grafting, twenty-two of the ninety-one tumors recurred. After primary resection, ten of the forty-four tumors recurred, whereas, after primary resection combined with bone-grafting, four of twenty-two tumors recurred. Primary amputation was performed for ten tumors none of which recurred. After primary irradiation, six of ten tumors recurred. Ninety-seven per cent of the recurrences occurred within two years of the time that the initial diagnosis was made. At the time of this study, 189 of the 218 patients were alive and well, and twenty-nine had died. Of the 189 who were alive and well, 121 had had no recurrence after primary treatment and sixty-eight had required secondary procedures. Of the twenty-nine patients who had died, fourteen had died of their tumor, three of postirradiation sarcoma, and twelve of unrelated disease. Eighteen of the twenty-nine dead patients had recurrent lesions. In the whole series, 180 secondary procedures were performed, 140 on sixty-eight patients who survived and forty on the twenty-nine patients who died. Of the 140 procedures on surviving patients, eighteen were for infection, five for lung metastases, four for fracture or non-union of a bone graft, and 113 for fifty-nine recurrences. The forty procedures done on the patients who died included one for infection, one for fracture of the bone graft, two for lung metastases, and thirty-six for recurrences. Of the seventy-seven patients who had recurrent lesions, fifty-six had one recurrence; sixteen, two recurrences; and four, three recurrences. Success in the treatment of recurrences was observed after a second curettage with or without bone-grafting in nine of sixteen patients, after secondary resection with or without bone-grafting in sixteen of thirty-five patients, after secondary irradiation in five of thirty-six patients, and after amputation in thirty-nine of forty-six patients. Irradiation, used for ten primary and thirty-six secondary tumors, was clearly beneficial in eight, possibly helpful in nine, and clearly ineffective in twenty-nine. Three of the patients so treated had postirradiation sarcoma. The average follow-up of patients who had radiotherapy was 9.2 years (the range four months to thirty-four years). In five of the six patients with pulmonary metastases, lobectomy was successful.