Sealing of the medullary cavity appears to be one of the first steps in the healing of a fracture. This may be produced by (1) a strand of periosteum, (2) a band of fibrin, or (3) a fibrinouslike plug produced by mass destruction of the red corpuscles. The seal is continuous across the cortex with the periosteum, and later becomes infiltrated by fibrous-tissue cells from this membrane. In other cases, however, these cells may arise from the marrow or develop directly from the lymphocytes caught in meshes of this fibrinous medullary cap.

By the second day, the periosteum from the site of fracture to the neck of the humerus is thickened. This would appear to be the result of a general stimulation rather than that caused by local trauma. By the third day, osteoid tissue has developed subperiosteally in regions well removed from the fracture. At the same time, osteoid tissue which has arisen from the endosteum, has also made its appearance at the fractured ends of the bone and, in conjunction with the original fibrinous plug, formed a more efficient seal for the medullary cavity. The development of this subperiosteal bone along the entire shaft was one of the most interesting features of this study.

The authors were not able to demonstrate any appreciable differences in the early stages of repair of bone in the three groups, but gained the impression that injections of estrogenic substance stimulated production of endosteal osteoid tissue. The insertion into the medullary cavity of a small plug of dense cortical bone as occurred in one of the cases could predispose to non-union.