when the tumor appeared in the wound. The capsule was divided and the tumor-substance appeared somewhat like fat. From its attachments it seemed futile to attempt its removal. The wound in the capsule was closed by catgut suture, and that in the abdominal wall by silk. The patient recovered from the operation and returned to his home June 22, 1889. Small portions of the tumor removed were examined and found to consist of fat-cells enclosed in a network of connective tissue. Numerous small, cellular elements were found in the stroma of the growth. The patient after his return home was relieved of his pain for a considerable period. The wound at the time of the operation did not close kindly, and later reopened, a portion of the tumor presenting in the wound as a fungoid mass. The relief from pain was undoubtedly due to the incision of the capsule, thereby relieving the tension. The growth continued to increase very decidedly, the legs became cedematous, respiration from the encroachment of the tumor became very much embarrassed, and finally, on October 21, 1889, the patient died.

Autopsy, held at Fonda, N. Y., twenty-seven hours after death. Body very much emaciated; rigor mortis well marked; thorax and abdomen only examined. Abdomen much enlarged, especially upon the right side. An incision from the intra-clavicular notch to pubes was made, passing in a curve to left of umbilicus. A second incision was made above the fungoid mass, separating it from the abdominal wall. The diaphragm was situated on the level with the fourth rib, on the right side, and not quite so high on the left. To the right of the vertebra the abdomen was filled by a large growth encapsulated and having the appearance of adipose. All the viscera were displaced to the left. The sternum was removed and both lungs found compressed, but otherwise healthy. There were no adhesions of the pleura. Each pleural cavity contained a small amount of serum. The more particular examination of the growth showed its origin to be posterior to the peritoneum and carrying it forward and inward. Externally it was continuous with the abdominal wall. To its median internal surface the cæcum was attached, the mesocæcum, if any, being obliterated. Superiorly the liver had been displaced to the left, and the right lobe was intimately adherent to the growth. The tumor was enucleated as follows:

The peritoneum along the inferior border was torn through and the tumor lifted up toward the thorax, dividing the peritoneum along the internal border of the growth. The right kidney was attached to the tumor and removed with it, after dividing ureter and renal vessels, which were much dilated. The renal vein, approaching the size of the vena cava after removal, lay completely posterior to the peritoneum. Liver smaller than normal from compression; gall-bladder empty and compressed; spleen normal in size, color, and consistence; left kidney slightly enlarged, but normal; right showed evidence of fatty degeneration. Omentum shrunken, free from fat; stomach dilated; intestines not distended and normal in appearance.

The growth (see Case II., Fig. 2) weighed fifty-six pounds, and later was divided in its length. Upon division a considerable amount of fluid escaped from numerous cysts. Examination of the cut surface showed many centres of softening, where the growth was myxomatous in structure. The external portion of the growth was of a firm consistence and resembled adipose in appearance. Portions of it were hardened in Müller's fluid and cut with freezing microtome. For this

part of the work I am greatly indebted to my assistant, Dr. Willis G. Macdonald.

Sections prepared from the outer portion were made up of adipose tissue, as were also other small areas found distributed throughout the growth. The softer portions were made up of a connective-tissue stroma, in the meshes of which were many multipolar cells, besides abundant small, round cells. The stroma was relatively very sparingly distrib

uted. In the growth we had, then, fat, myxomatous tissue, and small, round cells, the characteristics of lipoma, myxoma, and sarcoma. It may be termed an adipose tumor, presenting a condition of myxosarcoma, with sarcomatous infiltration.

CASE III.-I saw this case in consultation with Dr. H. E. Mereness, of Albany, N. Y., to whom I am indebted for the notes. Mr. N., aged thirty-nine, a brassworker by occupation, began to suffer from vague abdominal symptoms late in 1888, which were attributed by his physician to the poisonous effects of the metal with which he worked. There was a general failing in health, with loss of flesh, strength, and appetite. Early in June, 1889, he found himself no longer able to do his work and went into the country. While there he developed jaundice, and, upon examination, a tumor was found in the left hypochondriac and lumbar regions of the size of a child's head. He returned to the city, where I saw him and verified the foregoing history.

I found that he had suffered from no previous serious illness and had been of fairly good habits. The family history was free from tuberculosis. His father died of cancer of the stomach. He was very much emaciated, and the conjunctivæ were of a yellowish cast, which later deepened to a bronze. An exploration was advised, believing the growth to be connected with the left kidney, but refused by patient. The tumor continued to grow rapidly, and in the latter part of November, 1889, he died.

An abstract from the notes taken at the autopsy showed that the tumor sprang from the region of the left supra-renal capsule and in

volved it, as well as extending downward and surrounding the kidney. It presented itself forward, dislocating the pancreas and spleen, and by pressure impeded the flow of the biliary secretions into the intestinal canal. It was closely attached to the vena cava and aorta. It weighed upon removal six and one-eighth pounds, and upon section presented gross anatomical appearances, very similar to the case previously reported. Microscopically, there was a greater preponderance of sarcomatous and myxomatous elements. Chemical examination of the fluid removed from the open spaces showed a large proportion of mucin. (See Case III., Fig. 3.)

ANATOMICAL RELATIONS AND PATHOLOGY.

From the histories of these cases, it will be noticed that in two they took their origin from the capsule of the kidney, and in the other from the supra-renal capsule. A careful research of the reported cases, together with an examination of all accessible works on pathology, impresses one that the most frequent origin of these growths is in the connective tissue of the capsule of the kidney; the next most frequent seat being the supra, or impossible to determine with exactness.

Mr. Hulke, of Middlesex Hospital, reports a case of myxoma surrounding the left kidney, which he operated upon. A median section was made, when the tumor presented. The incision was continued

through the posterior blade of the peritoneum, just beyond the descending colon. The tumor proved to be a myxoma, and, although the patient recovered from the operation, it returned locally. The kidney was not involved by the tumor, and could have been enucleated.

The variety of growths embraced in this class present no less peculiarities than do the other features of their natural history. None can be said to be absolutely benign, even those which are made up entirely of histological elements, such as lipoma, fibroma, or myxoma, although they do not tend to the formation of metastasis, or to the infiltration of immediately surrounding tissues. It is true they show no great tendency to recurrence when completely removed, yet from the great size to which they develop, the consequent discomfort, and from their tendency to undergo degenerative changes, they cannot be classed as innocent growths.

As with any large abdominal growth, there is always a degree of encroachment upon the thorax and pressure upon other organs. In Case II., here reported, the inferior border of the liver was at the fourth intercostal space, and the cæcum and ascending colon were displaced toward the left side of the abdomen.

Those springing from the walls of the pelvis encroach upon or involve the bladder, the uterus, and rectum, very often presenting features which are exceedingly perplexing in diagnosis. The origin of a smaller number are reported respectively as from the retro-peritoneal lymphatics, the bodies of the vertebræ and bones of the pelvis, and the root of the mesentery. In by far the greater proportion of the cases no exact origin is given; indeed, from the subsequent changes in anatomical relations, it would seem quite impossible to determine the exact point of origin of many of the very large retro-peritoneal new-growths. They have almost uniformly presented themselves in the line of the least resistance, that is, anteriorly. In the first case which I report, the tumor presented in the back-a condition which can be attributed to its origin from the extreme upper border of the kidney, where it is less completely bound down by the lumbar muscles and fascia. When they have reached a sufficient size to attract the attention of the patient, they present themselves at either side of the umbilicus, somewhere in the region of the lateral lines drawn in the arbitrary divisions of the abdomen into regions, although they may appear centrally. With the growth viscera are displaced, not infrequently completely to the opposite side, although those springing from the left kidney may have the descending colon externally. Owing to circulatory disturbances within them, and their liability to a subsequent malignant infiltration, and from the development of cachectic conditions, they present, clinically, features which are almost immediately hazardous to life.

Cysts have been reported by various authors as arising from the retro

peritoneum. Czerny reports particularly a case of dermoid cyst springing from the capsule of the kidney. Others have reported cases of similar growths springing from the walls of the pelvis, particularly in the neighborhood of the sacrum. The so-called cyst of the broad ligament is very often partially retro-peritoneal in character.

Cysts in connection with the parenchyma of the kidney and pancreas, together with new-growths springing from the same source, are not considered in this paper, for the reason that they have already received at other hands sufficient consideration.

A careful study of retro-peritoneal tumors shows them to be of a mixed variety, containing the elements both of the lipoma and myxoma -tissues which are, histologically, very closely associated. Fat is developed from embryonal mucin, and in post-fœtal life occupies those spaces in the economy which later in the foetus are of myxomatous elements.

Again, in the retro-peritoneum both of these tissues are found, and doubtless many of these growths arise, under suitable conditions, from congenital neoplasms. There is, in these cases or classes of tumors, no inconsiderable amount of evidence to support the theory of Cohnheim relative to the origin of tumors. They are sometimes active in their growth, often become cystic, and reach immense proportions. Cases are reported of tumors of this character weighing eighty pounds. The distribution of the elements is very diverse. The fat may be regularly distributed throughout the tumor or may occur as islands located here and there. These growths are very often oedematous, and by chemical analysis present a large percentage of mucin. The microscope, besides revealing the usual elements of lipoma and myxoma, very often reveals a numerous round-celled infiltration in the stroma of the growth, pointing to a sarcomatous element in their character. While not presenting all the features of active malignancy that carcinoma and sarcoma do, yet they often show a marked tendency to recur locally when removed.

Clinical histories and pathological research both show that tumors containing embryonal elements are very apt to be malignant, and can very truly be said to be always so.

The tumors found in the kidneys of young children are, for the most part, mixed tumors, chiefly myo-sarcoma. Many are surely congenital, and are an example of a new-growth developing from embryonal tissue. They have by Grawitz recently been compared to a series of embryonal growths which spring from the supra-renal capsules, and have the appearance of adipose tissue, but are usually sarcomatous. (Orth, Pathologische Diagnostik.)

Sarcomas, either in typical form or in combination with myxomatous, lymphomatous, or fibromatous tissue, also occur, and have been observed quite often. Many of the tumors which have been described as sarcoma