1. Sixty-eight successive patients with tuberculous spondylitis were subjected to fusion procedures, homogenous bank bone being used. 2. There were 169 vertebral interspaces involved; 90 per cent. were between the fifth thoracic and the fifth lumbar vertebrae. 3. Exploratory operations were done between three and a half and ten months after the previous procedure in 101 cases. 4. The extenst of fusion to be attempted at any one procedure is dependent upon the condition of the patient, the availability of bone, and the skill of the operating surgeon. The goal for which we aim is solid fusion of all vertebral bodies involved in the abscess. 5. In 85 per cent. of the patients, final successful fusion, after repair of pseudarthrosis with bank bone, resulted. In 44 per cent. of these, successful fusion resulted after the primary procedure. 6. There was no relationship between the area most frequently diseased and the site most suceptible to pseudarthrosis. 7. The incidence of pseudarthrosis was greatest in the last three lumbar interspaces. In contrast to this, there was not a single instance of pseudarthrosis involving the interspace between the eighth and ninth thoracic vertebrae in twenty-four operations at this level. 8. Bank bone may be used in the repair of pseudarthrosis whether autogenous or homogenous bone was used in the primary fusion procedure. We believe cancellous bone, either autogenous (if available) or homogenous, is desirable. 9. Postoperative complications were relatively few and did not appear to be caused by the use of bank bone.