⬅ William McKinley: Autopsy Report

McKinley was shot twice at close range on Sept. 6, 1901. One bullet bounced harmless off his sternum and did not enter his body 1a. The other entered the left upper quadrant of his abdomen, piercing the front and back walls of the stomach 1b.

He underwent surgery within hours. He survived the operation, but died on the ninth post-operative day. Both his post-operative course MORE 2a and his autopsy SEE BELOW 2b have been meticulously documented 3 4.

There was intense controversy about McKinley's medical care MORE. Some thought that McKinley could have been saved had renowned surgeon Roswell Park performed the operation MORE. More recent commentators believe, however, that McKinley died from pancreatic necrosis, a condition which is still difficult to treat today, and which the surgeons of McKinley's time could not have treated or prevented 1.

Notes on the Autopsy on President McKinley September 14, 1901

Ordinary signs of death; ecchymosis in dependent portions of the body. Rigor mortis well marked. Upon the surface of the chest to the right of the midsternal line a spot 1 cm. in diameter, dark red in color, with a slight scab formation covering it; measuring from the supersternal notch the distance is 5 1/2 cm.; from the right nipple, 10 cm.; from the line of the right nipple, 8 1/4 cm. Surrounding this spot, at which point there is an evident dissolution of the continuity of the skin, is a discolored area of oval shape extending upward and to the right. In its greatest length it is 11 cm., and in its greatest width, 6 cm. It extends upward in the direction of the right shoulder. The skin within this area is discolored; greenish yellow and mottled. The surface of the abdomen is covered with a surgical dressing which extends down to the umbilicus and upward to just below the nipples. The innermost layer of cotton is covered or stained with balsam of Peru and blood. On removing this dressing a wound is exposed. Inserted in the wound are two layers of gauze, likewise impregnated with balsam of Peru. The wound has been packed with gauze saturated with the same substance. The wound is 14 1/2 cm. in length and is open down to the abdominal muscles. The layer of abdominal fat is 3 3/4 cm. in thickness. The appearance of the fat is good, a bright yellow in color. No evidence of necrosis or sloughing. In the left margin of the surgical wound, lying 1 cm. to the right of a line drawn from the umbilicus to the left nipple, 15 1/2 cm. from the nipple and 16 1/2 cm. from the umbilicus, is a partly healed indentation of the skin, and an excavation of the fat immediately beneath it. This extends down to the peritoneal surface. The base of the surgical wound is formed by folds of omentum. On making the median incision, starting from the supersternal notch and extending to a point just below the symphysis, the subcutaneous fat is exposed, which is of bright yellow color and normal appearance, except in an area which corresponds superficially to the area of discoloration described as surrounding the wound upon the chest wall. In this area the fat is of a red color, the connective tissue structure is infiltrated with dark red pigment. The subcutaneous fat is firm and measures 4 3/4 cm. in thickness. On opening the sheath of the right rectus muscle it is seen to be of dark red color. (Culture taken from ecchymotic tissue under the upper bullet hole and from between the folds of the small intestine.) (Three tubes from each locality on agar and gelatine.)

On opening the abdominal cavity the parietal surface of the peritoneum is exposed and is found to be covered with a slight amount of bloody fluid; is perfectly smooth and not injected. The great omentum extends downward to a point midway between the umbilicus and the symphysis. It is thick, firm; its inferior border is discolored by coming in contact with the intestines. Below the umbilicus a few folds of intestines are exposed. These are likewise covered with discolored blood, after the removal of which the peritoneal surface is found to be shiny. On the inner aspect of the abdominal wound the omentum is found to be slightly adherent to the parietal peritoneum, and can be readily separated with the hand from the edge of the wound. (Culture taken at this point, the surface of the wound.) At this point the omentum is somewhat injected. This adhesion to the omentum is found to extend entirely around the abdominal wound. The peritoneum immediately adjacent to the inner aspect of the abdominal wound is ecchymotic. In the omentum immediately beneath the abdominal wound is an incision, 5 cm. from the medial line and extending downward from the margin of the ribs 8 cm. On removing the subcutaneous fat and muscles from the thoracic wall, the point which marks the dissolution continuity of the skin upon the surface is found to lie directly over the margin of the sternum and to the right side between the second and third ribs. There is no evidence of ecchymosis or injury to the tissues or muscles beneath the subcutaneous fat. On making an incision through the subcutaneous fat directly through the wound upon the surface a small cavity is exposed about the size of a pea just beneath the skin, which is filled with fluid blood. (A section of tissue, including the lower half of the wound and extending through the subcutaneous fat, is taken for examination. The upper portion of the wound is removed for chemical examination.) The subcutaneous tissue underlying the area of discoloration on the surface of the chest wall shows hemorrhagic infiltration.

On removing the sternum the lungs do not extend far forward. A large amount of pericardial fat is exposed. Pleural surface on both sides is smooth. There are no adhesions on either side within the pleural cavities. The diaphragm on the right side extends upward to a point opposite the third rib in the mammary line. No perceptible amount of fluid in either pleural cavity. On opening the pericardial cavity the surface of the pericardium is found to be smooth and pale. The pericardium contains approximately 6 c.c. of straw-colored, slightly turbid fluid (some taken for examination).

On exposing the heart it is found covered with a well-developed paniculus. The heart measures from the base to the apex on the superficial aspect 10 1/2 cm. The right ventricle is apparently empty. The heart feels soft and flaccid. On opening the left ventricle a small amount of dark red blood is found. The muscle of the left ventricular wall is 1 1/2 cm. in thickness; dark reddish brown in color; presents a shiny surface. The average thickness of the pericardial fat is 3 1/2 mm. (Blood taken from the auricle for examination.) The left auricle contains but a small amount of dark, currant-colored blood. The mitral valve admits three fingers. The right ventricle, when incised in the anterior line, is found to be extremely soft; the muscular structure is 2 mm. in thickness. The paniculus measures 7 mm. The muscle is dark red in color; very shiny.

On opening the right auricle it is found to be filled and distended by a large currant-colored clot which extends into the vessels. The tricuspid orifice admits readily three fingers. The coronary arteries were patulous and soft; no evidence of thickening.

On unfolding the folds of intestine there is no evidence of adhesion until a point just beneath the mesocolon is reached, when, on removing a fold of small intestine a few spoonfuls of greenish gray thick fluid flows into the peritoneal cavity. On the anterior gastric wall is an area to which a fold of the omentum is lightly adherent. On breaking the adhesion there is found a wound about midway between the gastric orifices, 3 1/2 cm. in length, parallel with the greater curvature of the stomach, 1 1/2 cm. from the line of omental attachment. This wound is held intact by silk sutures. The cardiac end of the stomach is free. There is no evidence of adhesion at any other point on the anterior wall. The gastric wall surrounding the wound just mentioned, for a distance of 2 or 3 cm., is discolored, dark, greenish gray in appearance, and easily torn. On exposing the posterior wall of the stomach from above along the greater curvature of the stomach the omentum is found to be slightly adherent, a line of silk ligatures along the greater curvature of the stomach marking the site where the omentum had been removed. On throwing the omentum downward the posterior gastric wall is exposed. On the posterior wall of the stomach, a distance of 2 cm. from the line of omental attachment, is a wound approximately 2 cm. in length, held intact by silk sutures. The gastric wall surrounding this wound is discolored. On the surface of the mesocolon, which is posterior to the gastric wall at this point, is a corresponding area of discoloration, the portion coming directly in contact with the wound in the gastric wall being of dull gray color. The remainder of the surface of the posterior wall of the stomach is smooth and shiny. Beyond the surgical wound in the posterior wall of the stomach is found an opening in the retro-peritoneal fat large enough to admit two fingers. This opening communicates with a tract which extends downward and backward as far as the finger can reach. The tissues surrounding this tract are necrotic. On removing the descending portion of the colon a large irregular cavity is exposed, the walls of which are covered with gray, slimy material, and in which are found fragments of necrotic tissues. Just at the superior margin of the kidney is located a definite opening which forms the bottom of the tract traced from the stomach. On stripping the left kidney from its capsule, it is found that the superior portion of the capsule is continuous with the cavity. The weight of the left kidney is 5 oz., 1 gr. The kidney is readily stripped from its capsule; is dark red; the stellate veins are prominent; and along its greater curvature are numerous dark red depressions. On the superior aspect of the kidney is a protrusion of the cortex, dark-red in color, and in this protrusion is a laceration 2 cm. in length, extending across the superior border approximately at right angles to the periphery of the kidney and from before backward. On incising the kidney, the cortex and medulla are not easily distinguishable from one another; both are of rose-red color, the cortex measuring approximately 6 mm. in thickness. The vessels in the pyramids of farriem are very prominent. Beneath the protruding portion of the surface the cortex is dark red in color. This discoloration extends downward in pyramidal form into the medulla. The laceration of the surface marks the apex of the protrusion of the kidney substance. Between the spleen and the superior aspect of the kidney is a necrotic tract which extends down and backward and ends in a blind pocket. The tract, which includes the superior aspect of the kidney, can be traced into the perinephritic fat to a point just above the surface of the muscles of the back. The necrotic cavity, which connects the wound on the posterior wall of the stomach and the opening adjacent to the kidney capsule, is walled off by the mesocolon and is found to involve a considerable area of the pancreas. A careful examination of the tract leading down toward the dorsal muscles fails to reveal the presence of any foreign body. After passing into the fat the direct character of the tract ceases and its direction can be traced no farther. The adjoining fat and the muscles of the back were carefully palpated and incised without disclosing a wound or the presence of a foreign body. The diaphragm was carefully dissected away and the posterior portion of the thoracic wall likewise carefully examined. All fat and organs which were removed, including the intestines, were likewise examined and palpated without result.

The great amount of fat in the abdominal cavity and surrounding the kidney rendered the search extremely difficult.

The liver is dark red in appearance, the gall bladder distended. The organ was not removed.

The right kidney is embedded in a dense mass of fat; capsule strips freely; it weighs 5 ounces; measures 11 1/2 cm. substance is soft; cortex is 6 mm. in thickness. There are a few depressions of the surface, and the stellate veins are prominent.

The pancreas at its center forms part of the necrotic cavity. Through its body are found numerous minute hemorrhages and areas of gray softening, the size of a pea and smaller. These are less frequent in the head portion of the pancreas.

The cause of death having been established and the autopsy having lasted nearly four hours, it was discontinued, as a further search for the bullet could serve no useful purpose.

There is no evidence of organic disease in any organ examined.

* * * * * * * * *

In regard to the chemical and bacteriological report, I have just received, under date of September 24, from Dr. Matzinger, the following:

There was no bacterial or chemical source of poison found on either cartridge or weapon.

From the outer wound there was obtained a gas bacillus with the ordinary pyogenic organism, but no streptococcus.

The chemical examination of the necrotic cavity showed that the material was alkaline and no free hydrochloric acid; microscopically showed only tissue material that was disorganized and unrecognizable.

The rest of the work as it relates to the autopsy is still in an incomplete condition, and there is little prospect of arriving at any definite results within the very near future.

In Dr. Matzinger's report submitted later he says the clinical examination of the empty shells and cartridges were negative, and closes his report with the following paragraph:

The absence of known pathogenic bacteria, particularly in the necrotic cavity, warrants the conclusion that bacterial infection was not a factor in the production of the conditions found at the autopsy.

A copy of Dr. Gaylord's report, received by me on October 16, gives the following anatomical diagnosis:

"Gunshot wound of both walls of the stomach and the superior aspect of the left kidney; extensive necrosis of the substance of the pancreas; necrosis of the gastric wall in the neighborhood of both wounds; fatty degeneration, infiltration, and brown atrophy of the heart muscle; slight cloudy swelling of the epithelium of the kidneys.

"The piece of retro-peritoneal fat, where it forms part of the necrotic cavity, is seen on section to be covered with a thick gray deposit, which has an average thickness of from 4 to 6 mm. Beneath this, and separating it from the fat, is a well-defined area of hemorrhage from 1 to 2 mm. in thickness. The appearance of this piece of tissue is characteristic of the fat tissue surrounding the entire cavity. A section, made perpendicular to the surface and stained with haematoxylineosin, shows the infiltration between the fat cells or of fat necroses. The surface of the tissue which, in the microscopic specimen was covered by a layer of grayish material, proves, under low power, to consist of a partly organized fibrinous deposit. At the base of this deposit is evidence of an extensive hemorrhage, marked by deposits of pigment. The surface of the membrane is of rough and irregular appearance and contains a large number of round cells with deeply stained nuclei. Under high power the organization of the membrane may be traced from the base toward the surface. The portion immediately adjacent to the fat tissue consists of a network of fibrin inclosing large numbers of partly preserved red blood corpuscles. In many areas the red blood corpuscles are broken down and extensive deposits of pigment are found. Extending into the fibrin structure of the membrane are numerous typical fibroblasts and round cells. In some regions pigment is evidently deposited in the bodies of large branching and spindle cells. Here and there included in the membrane are the remains of fat cells, and toward the surface of the membrane a large number of round cells, scattered through the interstices of the membrane. There are but few polymorph nuclear leucocytes. Here and there in the membrane are fragments of isolated fibrous connective tissue, with irregular contours and an appearance suggesting that they are fragments of tissue which have been displaced by violence and included in the fibrin deposit. The fibrin in the superficial layers of the membrane is formed in hyaline clumps. The organization along the base of the deposit is comparatively uniform.

"Sections stained with methylene blue, carbo-thionin, and Gram's method, were carefully examined for the presence of bacteria, with negative results. Even upon the surface of the membrane there are no evidences of bacteria.

"The section of the left kidney, including the triangular area of hemorrhage described in the macroscopic specimen, reveals the following appearances; (Section hardened in formalin with haematoxylin-eosin). Examined macroscopically section represents a portion of kidney cortex made perpendicular to the surface of the cortex and including an area of hemorrhage into the substance of the cortex 1 cm. in length, measured from the capsular surface downward, and presenting a width of from 5 to 6 mm. The capsular surface has evidently been torn.

"Under low power the margins of the preparations are found to consist of well-preserved kidney structure. There is a slight amount of thickening of the interstitial tissue and occasional groups of tubules are affected by beginning cloudy swelling. The glomeruli are large and present a perfectly normal appearance. As we approach toward the center of the preparation occasional glomeruli are met with in which capillary loops are engorged, and the adjacent tubules contain red corpuscles. A short distance farther the kidney structure becomes entirely necrotic. Here and there the remains of tubules may be made out, and these are infiltrated with cells. The necrotic area presents a rough, net-like structure. As we approach toward the surface of the kidney we find that the necrosis becomes more marked. There is the merest suggestion of kidney structure, its place being taken by disintegrated red blood cells and leucocytes embedded in a well-defined fibrinous network. There is great distortion of the kidney structure about the periphery of the necrotic area. In this region a considerable amount of pigment is also found in the necrotic tissues.

"Under high power the characteristics of the necrotic tissues may be better observed. The kidney structure is broken up and torn into irregular fragments, infiltrated by red blood corpuscles and leucocytes. In the portion of the necrotic mass beneath the capsule the kidney structure is practically obliterated and is replaced by a network of fibrin which includes large numbers of red blood cells and leucocytes. Scattered through the entire necrotic area are frequent deposits of pigment. In the deeper portions of the necrotic area the margins of the fibrin deposit are invaded by fibro-blasts from the connective tissue structure of the kidney. The organization in these areas is, however, slight.

"Sections stained with methylene blue and Gram's method and carefully examined under oil immersion fail to reveal the presence of any organisms. In preparations stained with methylene blue the deposits of pigment may be readily observed. Section of the same tissue hardened in Hermann's solution and examined for fat shows the presence of numerous fat droplets, within the epithelium of the tubules which are adjacent to the area of necrosis. In the portions of the preparation more widely distant from the area of necrosis no fat is present.

"Section of the right kidney, hardened in formalin and stained with haematoxylin-eosin, reveals the presence of areas in which slight parenchymatous degeneration of the epithelium in the uriniferous tubules may be noted. These areas are not extensive, and are confined to single groups of tubules. The interstitial connective tissue of the organ seems to be slightly increased in amount but there is no well-defined round-celled infiltration. An occasional hyaline glomerulus is to be met within these cases, surrounded by increased connective tissue. The epithelium of the kidney tubules, aside from those in which the parenchymatous degeneration is present, is well preserved. The nuclei are well stained, protoplasm finely granular.

"A fragment of the stomach wall taken from the immediate neighborhood of the anterior wound, is in a condition of complete necrosis. The nuclei of the cells are scarcely demonstrable. The epithelial surface is recognized with difficulty. At its base are apparently a few round cells. Examination of the blood vessels reveals nothing characteristic. There is apparently no evidence of thrombosis. A section made through the gastric wall at some distance from the wound reveals the well-preserved muscular structure of the gastric wall, which presents no characteristic alterations. Superficial portions of the epithelium have apparently been affected by post-mortem digestion. However, in one portion of the preparation the epithelium is intact and shows distinct evidence of marked round-celled infiltration between the granular structures. The blood vessels contain red blood corpuscles with the usual number of leucocytes.

"The fragments of heart muscle which were removed from the right and left ventricular walls were examined in the fresh state and exhibited a well-defined fatty degeneration of the muscle fibers, and in the case of the right ventricular wall an extensive infiltration between the muscle fibers of fat was apparent. Sections from these fragments of muscle hardened in Hermann's solution are taken for examination. A fragment of muscle from the right ventricular wall was removed at a point where the fat penetrated deeply into the muscular structure, the ventricular wall at this point showing an average thickness of 2 1/2 millimeters. Under low power the muscle fibers are separated into bundles by masses and rows of deeply stained fat cells. The muscle fibers are seen to contain groups of dark-brown granules lying in the long axes of the cells. Under high power these are resolved into extensive groups of dark-brown pigment arranged around the nuclei. The muscle fibers are slender; the cross and longitudinal striation is well defined. Examined near the margin of the preparation, where the osmic acid fixation has been successful; all of the muscle fibers are found to contain minute black spherical bodies extending diffusely through all the muscle fibers about the entire margin of the preparation. These fine fat droplets are present in sufficient amount to speak of an extensive diffuse fatty degeneration of the muscle fibers. Where the large fat cells have separated the muscle fibers, these are found to be more atrophic than those in the central portions of the larger bundles.

"The examination of the section through the healed bullet wound on the chest wall reveals nothing of importance. The dissolution of continuity is filled in by granulation tissue, and there is evidence of beginning restoration of the epithelium from the margins. Stains for bacteria give negative results.

"In summing up the macroscopic and microscopic findings of the autopsy, the following may be stated. The original injuries to the stomach wall had been repaired by suture, and this repair seems to have been effective. The stitches were in place and the openings in the stomach wall effectually closed. Firm adhesions were formed both upon the anterior and posterior walls of the stomach, which reenforced these sutures. The necroses surrounding the wounds in the stomach do not seem to be the result of any well-defined cause. It is highly probable that they were practically terminal in their nature and that the condition developed as a result of lowered vitality. In this connection there is no evidence to indicate that the removal of the omentum from the greater curvature and the close proximity of both of these wounds to this point had any effect in bringing about the necrosis of the gastric wall, although circulatory disturbances may have been a factor. The fact that the necrotic tissue had not been affected by digestion strongly indicates that the necrosis was developed but shortly before death. The excavation in the fat behind the stomach must be largely attributed to the action of the missile. This may have been the result of unusual rotation of a nearly spent ball or the result of simple concussion from the ball passing into a mass of soft tissues. Such effects are not unknown. The fact that the ball grazed the superior aspect of the left kidney, as shown by the macroscopic investigation of that organ, indicates the direction of the missile, which passed in a line from the inferior border of the stomach to the tract in the fat immediately superior to the kidney. There was no evidence that the left adrenal gland was injured.

"The injury to the pancreas must be attributed to indirect rather than direct action of the missile. The fact that the wall of the cavity is lined by fibrin, well advanced in organization, indicates that the injury to the tissues was produced at the time of the shooting. The absence of bacteria from the tissues indicates that the wound was not infected at the time of the shooting and that the closure of the posterior gastric wound was effectual. The necrosis of the pancreas seems to us of great importance. The fact that there was no fat necroses in the neighborhood of this organ indicates that there was no leakage of pancreatic fluid into the surrounding tissues. It is possible that there was a leakage of pancreatic fluid into the cavity behind the stomach, as the contents of this cavity consisted of a thick, grayish fluid containing fragments of connective tissue. In this case the wall of fibrin would have been sufficient to prevent the pancreatic fluid from coming into contact with the adjacent fat. The extensive necrosis of the pancreas would seem to be an important factor in the cause of death, although it has never been definitely known how much destruction of this organ is necessary to produce death. There are experiments upon animals upon record in which the animals seem to have died as a result of not very extensive lesions of this organ. One experiment of this nature, reported by Flexnor, Journal of Experimental Medicine, Volume II, is of interest. The fact that concussions and slight injuries of the pancreas may be a factor in the development of necrosis is indicated by the researches of Chiari, Zeitschrift fur Heilkunde, Volume XVII, 1896, and Prager medicinische Wochenschrift, 1900, No. 14, who has observed (although a comparatively rare condition) extensive areas of softening and necrosis of the pancreas, especially of the posterior central portion, which lies directly over the bodies of the vertebrae, where the organ is most exposed to pressure or the effects of concussion. The wound in the kidney is of slight importance except as indicating the direction taken by the missile. The changes in the heart, as shown by the macroscopic inspection and the microscopic examination, indicate that the condition of this organ was an important factor. The extensive brown atrophy and diffuse fatty degeneration of the muscle, but especially the extent to which the pericardial fat had invaded the atrophic muscle fibers of the right ventricular wall, sufficiently explain the rapid pulse and lack of response of this organ to stimulation during life."

The cause of death of the President has been made plain by the autopsy. It was due primarily to a gunshot wound by a .32 caliber bullet fired at close range, devitalizing the tissues immediately surrounding its tract, so that gangrene of those parts injured, involving the stomach, pancreas, kidney, and other tissues were absorbed, and with the degenerated condition of the muscular tissue of the heart caused death, the final symptoms being those of exhaustion.

Report of Death

Name of deceased: McKinley, William.
Office: President of the United States.
Date of death: September 14, 1901.
Time of death: 2:15 a.m.
Place of death: 1168 Delaware avenue, Buffalo, N. Y.
Date of burial: September 19, 1901.
Place of burial: Canton, Ohio.
Cause of death: Gangrene of both walls of stomach and pancreas following gunshot wound.

I hereby certify that McKinley, William, President of the United States, died while at Buffalo, N. Y., as set forth in the record of his case, as follows:

The President was holding a public reception at the Academy of Music, Pan-American Exposition, Buffalo, N. Y., on September 6, 1901, and whilst shaking hands with the people was shot at 4:07 p.m. through the stomach by Leon F. Czolgosz.

There is good evidence that the disease (or injury) causing death was in line of duty, the facts being as follows: The President was shot by an assassin whilst receiving the people.

P. M. RIXEY, Medical Inspector, United States Navy.

I must mention here the giving up by Mr. John G. Milburn of his entire home in Buffalo and the devoting of his whole time and energy to the care of the President.

In concluding this report I must also refer to the untiring and devoted services of Mr. George B. Cortelyou, secretary to the President, who, with Mr. Nelson P. Webster, Mr. M. C. Latta, members of the Executive staff, and Mr. C. A. Conrad, of the Post Office Department, were on duty night and day. Executive Mansion Steward William Sinclair and Messengers Charles Tharin, Thomas Lightfoot, and Harry Mickie were also on duty at the Milburn house during the President's illness.

In obedience to the Department's orders, I was with the President's party at Buffalo, N. Y., on September 6. Upon arrival at the railroad station on its return from Niagara Falls, about 3:30 p.m., the President directed me to escort Mrs. McKinley to the Milburn house.

As soon as I learned of the attempt on the President's life, I hastened to his side at the Emergency Hospital on the Exposition grounds and was in the operating room with him at about 5:30 p.m. The President was under the influence of the anesthetic administered by Dr. Eugene Wasden of the United States Marine Hospital service. Dr. M. D. Mann, with a full corps of assistants, was ready to begin a laparotomy, which all deemed imperative.

Being satisfied with the completeness of the preparation and the ability of the operating surgeon, I made ready to assist and watched every step of the operation. The wounds having been closed, and the President's condition being good, I requested Dr. Roswell Park, the Medical director of the Pan-American Exposition, to send nurses and a surgical bed to the Milburn house and to take personal charge of the removal of the President, as I had to inform Mrs. McKinley of her husband's condition and make ready a room for his reception.

On his arrival I assumed charge of the case, having as consultants Dr. M. D. Mann, of Buffalo, N. Y.; Dr. Roswell Park, of Buffalo, N. Y.; Dr. Herman Mynter of Buffalo, N. Y.; Dr. Eugene Wasden, of the U. S. Marine Hospital Service. Dr. Charles McBirney, of New York, joined the consultations at 3 p.m. September 8, and left for home after the 9:30 a.m. bulletin of September 12. Dr. Charles G. Stockton, of Buffalo, N. Y., joined the consultations at 5 p.m. September 12. Dr. Edward G. Janeway, of New York, and Dr. W. W. Johnson, of Washington, D. C., arrived and Dr. McBirney returned after all hope had departed. All were present at the autopsy. Dr. H. G. Matzinger of Buffalo, N. Y., made all the urinalyses and also had charge of the chemical and bacteriological work. The histological examination of the tissues were made by Dr. H. R. Gaylord, who, with Dr. Matzinger, performed the autopsy.

In addition to the nurses mentioned, Miss Grace McKenzie, of Baltimore, Md., was employed after her arrival, and Miss Evelyn Hunt, of San Francisco, Cal., Mrs. McKinley's nurse, assisted as required.

## End ##