In conclusion, the following points should be emphasized:

1. Endothelial myeloma is a definite clinical entity and is probably of more frequent occurrence than is generally recognized. Many patients dying of prolonged sepsis from osteomyelitis are probably suffering from this tumor.

2. There is much difference of opinion as to the true nature of the tumor, and, until further proof is submitted, all theories should be considered.

3. The correct diagnosis can be made, with very rare exceptions, if due regard is given to an accurate history and other clinical tests. Erroneous operations for osteomyelitis are almost always performed after the occurrence of remittent attacks, and not in the initial attack.

4. There are three distinct and definite stages, as demonstrated by the roentgenogram, the knowledge of which should materially aid in diagnosis.

5. Radium and x-ray irradiation are the only known remedies that have a definite beneficial action on the evolutionary process; therefore, they should be employed in every case, before and after, or without radical operation. The tumor responds to both radium and x-ray, but the latter is of more value as it is more generally accessible. Irradiation as a diagnostic agent also is most valuable and should be more generally employed. The value of foreign-body protein, as Coley's toxins, is yet to be determined.

6. The mortality remains high, but has been reduced to some extent by utilizing more effectively our knowledge of the subject, for which gratitude should be expressed to the Registry of Bone Sarcoma.

7. Amputation or radical excision, preceded and followed by irradiation locally and to the chest, is apparently the treatment of choice. However, evidence as to the manner of treatment in the varying degrees of malignancy, and as regards age and location, is necessary before definite conclusions can be reached.

8. No effort has been made to discuss in detail the histogenesis or differential microscopic findings in bone tumors, as this would require one trained in this special field. The main objective of this paper is an analysis from a practical point of view of the salient features in a fairly large group of endothelial myelomata.