3o8 MANUAL OF BACTERIOLOGY. 



third week of the disease; it may be seen earlier; it is often delayed and ap- 

 pears late. The test frequently has to be repeated when the first result is 

 doubtful or negative. Reports indicate that the method is a great aid in the 

 diagnosis of typhoid fever, though not infallible. 



Considerable experience is necessary to acquire the judgment needed in 

 using 'this test. 



The agglutinating power becomes lessened after recovery, and usually is 

 wanting at the end of a year. Rarely it may be present for a longer time, a 

 fact that is to be borne in mind in making a diagnosis. 



Typhoid bacilli have frequently been obtained from the 

 stools of cases of the disease, but they are isolated only with 

 considerable difficulty. At autopsies they are best culti- 

 vated from the spleen, in which, however, it is to be remem- 

 bered, the Bacillus coH communis may also be present. Punc- 

 ture of the spleen with a sterilized hypodermic needle, during 

 hfe, has also been resorted to as a means of diagnosis. The 

 drop of fluid withdrawn may be examined by culture-methods 

 for typhoid bacilli. There is probably some danger to the 

 patient attending this procedure. Cultures made from the 

 blood, where several cubic centimeters are taken, show that a 

 few bacilli occur in the blood in a large proportion of cases 

 of the disease — ^probably in a majority. Typhoid badUi appear 

 in the urine in about 20 per cent, of all cases, and the exami- 

 nation of urine for them has been used in diagnosis. The 

 bacilli often occur in the gall-bladder. They have been found 

 associated with gall-stones, and have been supposed to be one 

 of the causes for the formation of gall-stones.* They may 

 remain present in the gall-bladder or in the urinef long after 

 convalescence from the disease. They have been demon- 

 strated in the "rose spots" on the abdomen. They may be 

 present in the lesions of the pneumonia, which frequently com- 

 plicates typhoid fever, and may appear in the sputum. 



Inoculation experiments in animals have not been very 



* Pratt. American Journal Medical Sciences. Vol. CXXII. igoi. 



t M. W. Richardson. Journal Experimental Medicine. Vol. IV. 1899. 



