for some reason or other have not healed. I have records of about five such cases: two are old museum cases, three operative cases; three of these involved the lower end of the femur, one the shaft, and one was situated in the upper end of the tibia.

Colvin, of St. Paul, was the first in this country to observe spontaneous healing. The cyst involved the upper end of the ulna. Dr. Colvin has sent me numerous X-rays of this case. It is now thirteen years since its first observation, the patient is nineteen years of age, the cyst has apparently healed with a slight residual expansion. Other cases are reported in the literature. In the majority the patients refused operation, and then further observation demonstrated the spontaneous healing. I have observed spontaneous healing in a small cyst in the lower third of the shaft of the femur (Fig. 18) in a child of nine; the cyst was discovered when an X-ray was made after a fracture. I now have a similar case under observation in which apparently spontaneous healing is taking place (Fig. 17).

Therefore, when we observe a young child with a small cyst not extending into the epiphysis or to the joint end, I think we may in the majority of cases delay and if future X-ray pictures show that the marrow shadow is becoming smaller, operation may be postponed with a large probability that it will be unnecessary.

However, when the patient comes under observation with an X-ray shadow of a larger marrow tumor nothing apparently is gained by delay. Operation apparently hastens healing.

Recurrences After Operations on Benign Bone Cysts.-In one case in which the cyst involved the upper half of the shaft of the humerus I subjected the young boy to three operations. At the third I transplanted into the defect a piece of the tibia. The cyst has now perfectly healed and the boy has unimpaired function. This patient came under my observation on each occasion before operation because of fractures. The interval was less than one year. When I compare this case with a similar one involving the upper half of the shaft of the humerus, I am confident that I misinterpreted the X-ray pictures after the second and third fracture as recurrences, when they were really only due to incomplete healing of the bone cyst. When the series of X-rays of my first cases subjected to three operations is compared with the one in which the cyst has ulti

FIT. 5.-Pathol. No. 10376. Giant-cell tumor of lower end of radius. X-ray of lower end of radius. Curetted by Dr. Chami ers, of Baltimore, in 1908. Well in 1919, eleven years, with perfect function. White female, aged twenty-five; sprain of wrist five months; pain and swelling two months later; swelling chief symptom. X ray by Dr. Cotton, of Baltimore. There was a distinct bony shell, and the tumor contained no cyst. Typical giant-cell tumor in the gross and microscopic specimens. This X-ray should be compared with Fig. 4, a benign bone cyst in the same region.

mately healed after one operation, the only difference is the evidence of refracture.

These two and subsequent observations of serial X-ray pictures demonstrate that it takes time for the bone cyst to heal -in the larger cases two to three or four years, but every X-ray shows some evidence of healing. Ultimately in the majority of cases the bone is restored to normal (Fig. 2), and one could not tell from the X-ray that the shaft had ever been the site of a cyst. For this reason it is my conclusion that the operations for the so-called recurrences recorded in the literature and described to me by colleagues have, as in my case, been unnecessary. Surgeons and roentgenologists should by repeated X-ray plates familiarize themselves with the normal healing of a bone cyst, whether it has been operated on or not. Refracture is not a sign of recurrence.

Diagnosis of the Benign Bone Cyst.-The most helpful clinical fact is the youthful age of the patient. With few exceptions the age of the individual with the giant-cell tumor is between twenty-five and forty, and with the malignant bone cyst over forty. The location of the marrow shadow can also to a large extent be depended upon: the benign cyst is as a rule on the diaphyseal side of the epiphysis; the giant-cell tumor and the malignant bone cyst with few exceptions, although they may involve the shaft, extend through the epiphyseal line into the epiphysis and usually destroy the cancellous bone leaving only the thin bone-joint capsule.

Pain and tenderness are not marked features of the bone cyst. The majority of patients are not aware of the presence of the cyst until fracture takes place and an X-ray picture is made. In the giant-cell tumor swelling of the lower end of a bone without very much pain or tenderness is the prominent clinical feature, and usually there is a history of injury, rarely with fracture. Unfortunately in all of the cases under my observation and reported in the literature no X-rays have been taken after the trauma and before the appearance of the swelling, so that we have no evidence of the actual condition of the involved bone at the time of the injury and no proof that trauma is the cause of the disease. One, however, gets the impression that trauma is a factor in the causation of a giantcell tumor, and not so in the benign bone cyst.

In the malignant bone cyst extreme pain and tenderness are marked clinical features before the appearance of any swelling,

FIG. 6.-Pathol. No. 10975. Giant-cell tumor of lower end of radius. X-ray of a central giant-cell tumor of the upper end of the tibia producing a huge expansion and involving the epiphysis to the joint. This X-ray should be compared with Figs. 8 and 9, which illustrate the difficulty of distinguishing the benign bone cyst from the giantcell tumor. Operation by Bloodgood in 1910-resection with bone transplantation. The patient is well in 1919, nine years since operation. The functional result is not good-the patient walks with an apparatus, but without crutches. The patient is a white female,

aged 26; pain in the head of the til ia three and one-half years; X-ray two years ago showing small marrow shadow in upper end of the til ia the size of a twenty-five-cent piece. Diagnosed tuberculosis and operation advised. This X-ray was lost. Had this X-ray been preserved, it would have presented the earliest picture of a central giant-cell tumor on record. After this examination swelling increased and the X-ray reproduced here is the one taken before operation. The bony shell is practically gone, making curetting impossible. If this patient had been operated on earlier curetting could have been dore and she would now have a limb with perfect function. I reported such a case in the Johns Hopkins Hospital Bulletin, May, 1903, xvi, ard in the Annals of Surgery, August, 1910. The giant-cell tumor in the upper end of the tibia was curetted in 1902, and the patient is well in 1919, seventeen years, with perfect function.

and in this group there is usually a history of trauma. Pathological fracture is rare, perhaps due to the fact that the pain and tenderness cause the individual to avoid trauma.

Central Giant-Cell Tumor.-In discussing this group of tumors under Benign and Malignant Bone Cysts of the Long Pipe Bones I do not wish to convey that cysts are common in the giant-cell tumor. As a matter of fact, they are very rare. But that they may occur must be borne in mind in the differential diagnosis.

We know that when we explore a bone cyst with or without a connective-tissue lining, it makes very little difference whether one cleanses the bone cavity of its connective-tissue lining or not, nor what is done to the cavity when there is no lining. We know that the cyst has a tendency to spontaneous healing. Complete removal of the ostitis fibrosa tissue is unnecessary and would only be possible by subperiosteal resection, because the Haversian canals of the bony shell are filled with the same inflammatory tissue as that lining the cyst, or filling the marrow cavity. However, in the giant-cell tumor all of this tissue must be removed, either by curetting thoroughly the bony shell, followed by thorough cauterization with some chemical or, in more extensive cases, when the bony shell is destroyed, by resection, while in the malignant bone cyst the bony shell with its periosteum must be removed either by resection or amputation. Differential diagnosis, therefore, of cysts in the marrow cavity of the long pipe bones is essential, unless one chooses to amputate or resect in every case. Since, as I believe, the contents of the benign bone cyst is never hemorrhagic or bloody, as it is in few cases of giant-cell tumor and in most of the cases of malignant bone cysts, the finding of such hemorrhagic contents of the cyst should immediately exclude the bone cyst