FIG. 9, CASE III.-Photograph of patient lying on her back;

left convex scoliosis; before treatment.

of the chest. The elements in this thorax are similar in shape to Figure 6. See how Figure 6 corresponds to the dotted lines in Figure 2. Figure 6 is exaggerated.

Upon examination it was found that with each inspiration the boy's chest over the left anterior half expanded. more than the right anterior half, and more over the right posterior half than over the left posterior half.

During the three years the deformity was developing

the posture, dependent in this case upon a short leg plus "skeletal insufficiency," caused the respiration after the manner indicated to force out the chest where the respiration was increased, and the chest to flatten where the respiration was decreased. This agrees with the observations on the shape of parts of the chest influenced by decreased respiration in one part and increased respiration in another part.

If the individual by bending to the right assumes a left convex scoliotic posture, the result upon the respiratory expansion is the opposite of that produced by the right convex scoliotic position. In this instance the right anterior half of the chest expands more than normal, the left anterior half less than normal, the right posterior half less than normal, the left posterior half more than normal. The effect upon each element then is as illustrated by the dotted line in Figure 7.

Figure 8, Case III, shows the prominence of the ribs in the left posterior half and the flatness of the right posterior half, while Figure 9 shows the same patient lying on her back. At a, the right anterior half, the chest is more prominent than the left anterior half. The patient's left costal border was more prominent than the right costal border. The latter fact is more apparent in the photograph, which was recently taken. No photograph of this patient in this position was taken before treatment was begun. An element in this case in the upper part of the chest had the shape shown in Figure

IO.

See how this corresponds to the dotted lines in Figure 7. Examination of this patient's respiration showed increased expansion over the upper part of the right anterior and left posterior halves and decreased expansion over the left anterior and right posterior halves, no mat

before treatment.

FIG. 14, CASE I.-With jacket on; back view. FIG. 15, CASE I.-With jacket on; front view. FIG. 16, CASE I.-With jacket on side view

ter what the posture she might assume the type of scoliosis was not changed.

It is plain from the foregoing illustrations and statements of fact that the respiratory expansion forces induced by the scoliotic posture, no matter what may be the original cause of the spine bending, tend to produce the same type of chest deformity as the rotary force, are in themselves rotary forces, and add to the torsion of the spine while their predominance in one part of the chest materially assists in forming a prominence of the thoracic wall, and in another part their relative weakness permits the chest to flatten. As the shape of the ribs and therefore the shape of the thorax is dependent upon the foregoing, it is plain that these forces must be continuously opposed in the treatment of the bone deformity.

Cases I, II, and III will illustrate the result upon them thus far from acting upon these principles.

Case I has been under treatment by correction jackets since May 1, 1906.

Case II was under treatment by correction jackets for fourteen months. He was a carpenter. Nine months ago he left the city and has not returned. Circumstances prevented his receiving gymnastic treatment subsequent to the use of the jackets. I have written several letters to him, but have received no reply.

Case III has been under treatment intermittently for three years. Twice she went for six months without a jacket, several times she has gone for one and two months without a jacket. In order not to have too many illustrations, only these cases of extreme deformity are reported in this article.

Figure 3. Case I, is a photograph of the back when treatment was begun, May, 1906. same view November 26, 1906.

Figure II shows the Figure 4, taken with

f