amount and ceases in about a week or ten days. One ear is usually involved, but occasionally both may be discharging at the same time.

The interference with the hearing is usually not very great, and entirely clears up in a few weeks or months.

As before stated, when first observed in the pus, the organism appears to be in pairs, but further observation showed the organism growing in short chains of four or more cocci, and occasionally as many as twelve, or more. When grown on culture media, however, they produce chains of greater length. On solid media they produced small pearly pin-point colonies, and at 37 degrces Centigrade reached their maximum growth in about twenty-four hours. In a few days the colonies appear ragged at the edges and lose their pearly appearance. It grows at the ordinary room temperature, and does not die out quickly.

The organism appears flattened on both sides, and has very little, if any, tendency to group itself in pairs in the chain formation. Occasionally one or two will appear with a greater diameter than the adjacent cocci. The organism decolorizes by Gram's method of staining, but stains readily by ordinary dyes. In bouillon there is no diffuse cloudiness, but flocculent masses collect in the bottom of the tube and occasionally on the sides. Microscopical examination shows these masses to consist of a network of streptococci.

At first this organism was supposed to be the streptococcus pyogenes. However, its staining reaction did not appear to conform to this organism. Quite a variety of streptococci have been described from time to time by various investigators, and attempts have been made to classify these different streptococci. It is generally considered, however, that these streptococci found about the human body are types of this streptococcus pyogenes. Ir

1891 Von Lingelshein classified these streptococci as pathogenic and non-pathogenic. Under the pathogenic, he classified all pyogenic streptococci which grew into long chains and spoke of them as the streptococcus longus, and under this head he sub-classified the streptococcus murisepticus, and the streptococcus pyogenes, and also the streptococcus erysipelatos. Under the head of the non-pathogenic he classified the streptococcus brevis. In classifying these he speaks as follows:

"These can not be distinguished one from the other in cultures in highly albuminous media (pus, blood-serum), but present constant differences when cultivated in bouillon. The decisive characteristics in this medium are: macroscopic, the cloudiness of the medium; microscopic, the length of the chains."

Since this classification some investigators describe streptococci which conform to Von Lingelshein's streptococcus brevis, but they call the organism streptococcus pyogenes.

Goadby describes the streptococcus brevis which he finds constantly in the mouths of people, both in health and disease. While in the mouth this organism appears usually in diplococcic form, but grown in chains on culture medium. This streptococcus brevis resembles closely the organism which we have isolated from these cases of suppuration of the middle-ear. With the exception of slight variations in growth on culture media it appears identical, except that the organism that we have isolated does not stain by Gram.

Von Lingelshein also states that the brevis uniformly clouds the bouillon; the organism we have isolated leaves the bouillon clear, forming a deposit in the bottom of the tube, thus conforming to his description of the growth of the streptococcus longus.

From time to time various investigators have isolated

streptococci from cases of otitis media. Levy and Schrader found a staphylococcus albus in pure cultures three cases out of ten in which paracentesis was performed. Netter found streptococcus pyogenes thirteen times in eighteen cases. Scheibe obtained the streptococcus pyogenes, diplococcus pneumoniæ, staphylococcus aurius, etc. Bordoni-Uffrenduzzi and Gradenigro found the diplococcus pneumoniæ to be the most common organism in middle-ear suppuration.

The fact that we have found only one organism in suppuration of the middle-ear is probably a coincidence, but that we have found the same organism in so many cases indicates that it is the most common factor in the production of otitis media in our section of the country.

THE TREATMENT OF THE BONE DEFORMITY OF THE THORAX AND SPINE IN SCOLIOSIS BY PLASTER JACKETS, UTILIZING THE EXPANSION OF THE LUNGS AS A CORRECTIVE FORCE.

BY MICHAEL HOKE, M.D., AND C. R. ANDREWS, M.D., ATLANTA, GA.

In November, 1903,1 I (Dr. Hoke) reported the case of a young lady, upon whose ribs I had operated in order to correct a very bad deformity of the chest-the result of scoliosis. Previous to the operations she had had gymnastic treatment for a year and a half. This treatment had made her more flexible, but the deformity of the bones was as pronounced at the end of the treatment as it was when the treatment was begun. A number of attempts were made to conceal the deformity without success. At that time there was no known method by which the degree of bone deformity which she had could be corrected. The operations were therefore devised and performed. The details of the operations will be found in the reference. Considering all the accompanying circumstances I may say the result is satisfactory. The patient presents no appearance of deformity with her clothes on without the use of pads, but she wears a celluloid corset. Her spine is still as crooked as it was before the operations were done, but the lines of the thorax from the armpit down are almost symmetrical. There is no prominence of either hip, the shoulders are in the same plane with the pelvis, and neither shoulder blade protrudes. Her circumstances were such that she was never

I The American Journal of Orthopedic Surgery, November, 1903. 2 A Study of a case of Lateral Curvature of the Spine. A Report on an Operation for the Deformity.

able to remain in the city for postoperative exercises and massage. I may assert with accuracy that at the time. these operations were done upon this patient it was impossible to accomplish the result obtained by any other method of treatment than the operation performed.

It occurred to me after the operation upon this patient, when she returned for a jacket, that the expansive power of the lungs might be used in these cases as a corrective force to fill out the flat areas, while the prominence of the chest were being prevented from expanding; hence the purpose was formed to try this experiment upon some cases as badly deformed as the case operated on in order to see if a similar result might be obtained by this simpler method. Thus no other patients were operated upon. While there remained a great deal of detail work to be done the progress made so far in this method warrants this preliminary report.

Figure is a photograph of two complete "thoracic elements," the fourth and fifth taken from a cadaver selected at random, the lungs of which were normal. The term "complete thoracic element" is used to represent the bone ring formed by a vertebra, the attached ribs and the coextensive part of the sternum.

Figure 1 shows how symmetrical is the normal thoracic element. It is evident that in the growth and development in life of the thorax of which the element was a part, equal forces were at work at corresponding areas symmetrically expanding the element. These forces were the respiratory forces of expansion. Given normal lung tissue in both lungs, man's natural posture and no pressure upon the chest wall, the corresponding areas of the chest wall are subject to the same amount of expansive lung force eighteen times per minute. The bones of each