THE RAPID development of extradural spinal cord compression due to tumor demands immediate action when diagnosed clinically. The variety of neoplasms capable of producing this syndrome in different settings makes organizing an investigative program an impossible task if different therapeutic trials are to be tested for efficacy. The causative agent of compression may vary from a highly radiosensitive lymphoma to a radioresistant terato-carcinoma of the testis. The onset may be acute and dramatic or insidious and chronic. The principles of management are often determined by the first physician to encounter the patient and may reflect his specialized knowledge of neurosurgery, radiation therapy, or chemotherapy rather than a consistent prescription or philosophy. The emergency presentation of such a patient and the disastrous consequences of unsuccessful treatment have led to the use of combination therapy, making it difficult to determine the contribution of each modality to success or failure in management. The limited and varied clinical experience has served as a stimulus for the development of an animal model. Because of the apparent parallel to superior vena cava obstruction, we have been working with an experimental system using the same tumor and rat line as in our previous studies (1). EXPERIMENT I: Production of Spinal Cord Compression Syndrome (10 animals): In order to simulate the human problem of spinal cord compression by epidural invasion of lymphoma, an animal model was developed employing 4-month-old Sprague-Dawley rats and a transplantable Murphy-Sturm lymphoma. The technic was similar to that described for the production of superior vena caval obstruction in the same rat line (1). A tumor was excised from the flank of a rat and cut into fine pieces under sterile conditions. The pieces were sucked into a trocar, and one fragment was transplanted into the nape of the neck of each rat (Fig. 1, A). A palpable mass developed in one to two weeks, causing flexion of the head (Fig. 1, B). The tumor invaded the epidural space which exists between the base of the skull and the cervical vertebrae in rats and opens with neck flexion. With compression of the spinal cord, a consistent, predictable, and measurable pattern of neurological loss was produced. The grading system consists of: 1+ Paresis of one upper extremity—the animal appears to be unsteady in gait and stumbles on one side 2+ Paralysis of one upper extremity—the animal walks but falls to one side and does not move the limb 3+ Paralysis of both upper extremities—the animal moves 4+ Paralysis of upper and lower extremities—the animal is immobilized. The pattern of progression was recorded by motion pictures for detailed study and analysis.