intendent being a physician, in many of the hospitals that are preeminent, those in Cleveland, Chicago, and Philadelphia, the superintendents are laymen. But they know their business and know how to distinguish between medical treatment and hospital business.

In reply to the remarks of the third gentleman, just as soon as a community hospital ceases to maintain wards for the charity patients it loses its purpose. It is to be hoped that our hospital will not come to that.

As to Dr. Cook's statement that he visited his patients four or five times a day; he would not have been able to do that had he not spent most of his time in the hospital. For that very reason I hope the time will come when every man who is privileged to practice in our hospital will be able to have his office adjacent to, and even within, the hospital walls so that much of his time will be spent in the hospital. The old "walking of the wards," formerly so frequent in England, was responsible for the high medical character of the men of that day. The "walking of the wards" of our hospitals is something that we have let go by the board, with a loss.

As to whether our men do the better for the patients with the group practice idea; there has been nothing in the discussion that I would take exception to. It is all perfectly true. There are several points of view in group practice, however. Great progress is being made in group practice for the reason that it is but another name for co-operation. It is a process by which co-operation is made more sure and scientific. It is a method rather than a principle. Is there anything so greatly objectionable about the matter?

Outline of the Superior Strait of the Pelvis

by Means of the X-Ray.*

DR. HERBERT THOMS, New Haven.

I wish to present for your consideration this afternoon a brief outline of a method for producing an outline of the superior strait of the female pelvis. This method has been developed during the past year at the Grace Hospital, New Haven, and the results have been so striking that I feel that it may become an important adjunct to obstetric procedure. When we consider the great advances that have been made in roentgenology it is at first glance somewhat surprising that no simple and accurate method exists for outlining the inlet of the female pelvis. True it is that some idea of the shape of this bony ring may be obtained, but the size is erroneous and from the malformed and enlarged image that ordinarily results, no accurate information may be obtained.

The chief causes of the distortion thus produced are two-fold. First, the divergence of the rays from the target or tube, and secondly, the distance of the part to be measured or photographed above the sensitive plate. I shall not take the time to go into the historical aspect of this problem. This I have published in another place in a paper dealing with this same subject. The most important plane of the pelvis from the obstetrical point of view is the superior strait or pelvic inlet, and the methods of pelvimetry that are now in general use, particularly the measuring of the diagonal and external conjugate diameters, give us at best only an approximate measure of but one diameter of this plane, the anteroposterior.

If then it becomes possible in cases where a contraction or deformity of the inlet is suspected for us to outline a diagram of the pelvic inlet in its true proportions, it is at once obvious that

*This paper was read in full at the Annual Meeting of the New York Obstetrical Society on May 9, 1922, and is published in full in the Proceedings of that Society in the American Journal of Obstetrics and Gynecology.

this becomes a valuable aid. The chief difficulty to overcome is that of the distortion which takes place because of the distance of the part to be measured from the sensitive plate. This is caused by the divergence of the rays from the target to the plate (see Figure 1). If we increase the distance between the target and the plate the divergence of the rays and the distortion produced become less, the rays which come through the pelvic ring becoming more perpendicular or normal. It is obvious that in order that the distortions thus produced may be equal in all directions it is essential that the target shall be placed over the center of the superior strait and that all portions of the plane of the strait shall be equidistant from the sensitive plate. This is found possible when the patient is in the semi-recumbent position, with the back arched in a manner which I will describe. In order to make certain that the promontory of the sacrum and the symphysis are in the same horizontal plane a Martin's pelvimeter is placed upon the patient in the same position as that used to determine the external conjugate or Baudeloque's diameter. One point of the pelvimeter rests in the depression on the spine of the last lumbar vertebra, the other rests on the skin over the upper and anterior margin of the symphysis pubis.

Previous workers in this field have stated that it is impossible to bring the plane of the superior strait in the living subject parallel with the horizontal. When the patient is in the ordinary semi-recumbent position with the pelvimeter applied it will be noted that the posterior point of the instrument is lower than the anterior and the plane of the superior strait is not parallel with the horizontal or sensitive plate. If, however, before the exposure is made the patient is asked to arch her back more or less as will be found necessary, it will be seen that both points of the pelvimeter may be brought equidistant from the sensitive plate. Arching the back furthermore tends to throw the upper part of the trunk away from the vertical. It is my custom to have the pelvimeter held by myself or an assistant in position while the exposure is made. This for two reasons. First, because it can be better maintained without artificial support, and secondly, to better reassure the

P' sy'

T

PR

PR'

P'

FIG. I. T is the target or tube. The broken line represents the plane of the Superior Strait and P the ends of the pelvimeter, SY the Symphysis and PR the Promontory of the Sacrum. It is obvious that these points will be projected to the sensitive plate at the points P' SY' PR' P' if the plane of the Superior Strait is parallel with the sensitive plate and that these points will bear the same relation to each other as P SY PR P. In other words the image will be enlarged equally in all directions.

patient that no harm will come from the formidable looking apparatus overhead.

The target is placed between three and four feet from the plate and a Bucky diaphragm used because much clearer negatives are produced in this manner. The exposure is made in the usual manner and the patient holds her breath during this period. Dr. Louis H. Wheatley, roentgenologist at Grace Hospital, has worked out the exposures and other points of roentgenologic technique with which I am unfamiliar and inasmuch as he is. to open the discussion he will undoubtedly speak of these things. It is obvious that the penetration of the rays directed toward the posterior or sacral part must be considerable in order to register satisfactorily.

The plate or film is developed, dried, and viewed. It will be noticed that not only is the pelvic ring enlarged equally in all directions but also the ends of the pelvimeter are shown farther apart than the actual distance measured on the patient.

It is apparent that the ratio between the actual distance of the points of the pelvimeter and that measured between the photographic images of these points represents the ratio between the actual size of the inlet and the increased size as shown upon the negative. With this knowledge the question of mensuration becomes one simple proportion.

In order to obtain a more graphic and permanent record a picture of the outline of the superior strait in its true proportions is produced, a positive image of this negative reduced in such extent that all distortion is corrected and the superior strait is represented as it actually is. This is accomplished by means of the camera. I have used an 8 x 10 view camera with a good steady tripod. The negative is placed in the X-ray view box or other illuminating apparatus and a picture made on a plate or film. In focusing the reduction must be made so that the ends of the pelvimeter shadows in the negative shall be the same distance apart on the ground glass as when measured on the patient. Thus if the pelvimeter read 20 cm. on the patient at the time of exposure and in the negative measured 24 cm. these points must be 20 cm. on the ground glass when focusing so that in the positive picture