The clinicopathologic features of 53 cases of postradiation soft tissue sarcoma (PRS) were correlated with the physical characteristics of the administered radiation. All but three patients received radiation for malignant processes. Of the secondary sarcomas, malignant fibrous histiocytoma (MFH) accounted for 36 cases (68%), followed by seven extraskeletal osteosarcomas (13%), six fibrosarcomas (ll%), two malignant Schwannomas (4%), one extraskeletal chondrosarcoma, and one angiosarcoma. The sex incidence, age of the patient at time of diagnosis, and location of the PRS correlated only with the clinical characteristics of the initial treated condition. The latency period (mean 10 years) showed an indefinite relationship to patient survival but no definite relationship to the patient's age at the time of the initial radiation. There was no difference between patients treated with megavoltage radiation (39 patients) and with or-thovoltage radiation (seven patients) in the type of sarcoma, location, or survival, although the orthovoltage group received a lower mean radiation dose (3880 rads) than the megavoltage group (4446 rads). Mega-voltage radiation, however, produced deeper tissue radiation changes and was associated with a shorter latency period. Most PRS were poorly differentiated, produced abundant collagen, and had a dismal prognosis. Cancer 622330-2340,1988. HERAPEUTIC RADIATION has long been recognized T as an inducing agent in the development of malignant neoplasms ever since 1902 when Frieben' reported a case of squamous cell carcinoma arising in the hand of an x-ray technician. Previous reports dealt primarily with postradiation carcinomas or bone tumors, but rarely with soft tissue sarcomas. However, as early as 1904, Perthes2 described a spindle cell sarcoma that followed radiation therapy for lupus. In 1907, Coenen' concluded that 12% of radiation-induced tumors were sarcomas, and one year later, Marie, Clunet, and Ravlot-LaPointe4 provided experimental proof by inducing sarcomas in irradiated rats. To date, only a few reports on postradiation sarcomas have attempted correlating the various radiobiologic parameters used in clinical practice with the development and histopathology of soft tissue ' The reasons for such a dearth of useful correlative information stem from the long latency period experienced with most post