In a retrospective study of 154 congenital dislocations of the hip treated by closed reduction with and without preliminary traction at Gillette Hospital during the period January 1, 1948, through December 31, 1967, the incidence of avascular necrosis was analyzed after follow-ups ranging from two to more than nineteen years. There were twenty-seven hips with complete necrosis and twenty-four with partial necrosis. Eight of the normal hips had shown Type-I changes previously described by Salter, Kostuik, and Dallas. An attempt was made to identify the factors responsible for avascular necrosis of the capital femoral epiphysis, either complete or partial. As a result of this study the following conclusions were reached: 1. The older the child is, the more frequent the complication of avascular necrosis, all other factors being equal. 2. There is a direct correlation between inadequate traction and the incidence of avascular necrosis of the femoral head. 3. There is a direct correlation between the Lorenz position and the incidence of avascular necrosis of the femoral head. 4. Avascular necrosis usually results in permanent deformity of the femoral head, but there is a spectrum of deformities which in all likelihood depend on the severity of the vascular insult. 5. A program of adequate traction, gentle reduction, and avoidance of the extreme Lorenz position reduced the incidence of total avascular necrosis from 34.8 per cent during the first five years of the study period to 4.5 per cent in the last five years. The incidence of partial necrosis, on the other hand, remained essentially the same.