15 



ing rise and evening fall. A small pus sac was found posterior to 

 the olecranon of the left arm. Aspiration yielded a few drops of 

 lemon-colored serous liquid which, microscopically, contained 

 numerous pus cells and a few capsulated rod-like bacteria. Later 

 a larger pus sac was found on the dorsum of the left foot. 



Sputum examination. May 30, mucopurulent, containing a few 

 blood cells; no tubercle bacilli. 



Blood examination. May 29, 21,000 leucocytes; June 3, 23,400 

 leucocytes; no parasites. 



Urinalysis. May 29, 1,021; alkaline, trace of albumin; no casts, 

 many leucocytes. June 2, 1,019, alkaline, no albumin; no casts, 

 few leucocytes; bile. 



Diagnosis. Pyemia; acute articular rheumatism, both elbows, 

 knees, and ankles; suppurative inflammation of left elbow (type 

 undetermined). 



Treatment. The treatment was essentially supporting and pal- 

 liative. 



About June 6 he developed a cutaneous eruption which was con- 

 sidered to be that of smallpox, since a smallpox patient had been 

 removed from an adjoining bed "seven days before the eruption 

 appeared on Marshall. He was removed to the military smallpox 

 hospital and died on the following morning. The body was sent 

 to the city morgue marked "suspected smallpox case." 



Incomplete autopsy No. 987 (about twenty-four hours after 

 death). No autopsy was requested, but Dr. W. E. Musgrave and 

 Dr. W. R. Brinckerhoff, who happened to be at the morgue, noticed 

 the peculiar cutaneous eruption and brought pieces of the skin to 

 the laboratory for further examination. Large areas of skin sur- 

 face were covered with the numerous, closely set vesicles of miliaria, 

 and among these were numerous pustules of varying size. Some 

 of these pustules were capped by a vesicle, which in some instances 

 showed a central depression giving an appearance of umbilication. 



None of us thought of human glanders until a miser oscopic 

 examination showed that the pustules contained numerous bacteria 

 morphologically resembling Bacterium mallei rods about the 

 length of and somewhat thicker than the tubercle bacillus, which 

 stained irregularly with Loffler's methylene blue and lost the stain 

 in Gram's method. (See PI. IV, fig. 2.) No such bacteria were 

 found in smears made from the abscess in the left ankle. 



Bacteriologic examination. Cultures were made on -f- 1 glyc- 



