ip-L?] FATE OF TUBERCLE BACILLI OUTSIDE THE ANIMAL BODY 343 



TESTING THE Nineteen tests from these samples were made for 

 SAMPLES the presence and for the virulence of tubercle 



bacilli. These tests were made on the 9th, 23d, 

 and 44th days, and thereafter about once a month until the last 

 test, which was made December 5, 1911, 586 days from the begin- 

 ning. The sample of the bovine tubercle bacilli kept in the cotton- 

 stoppered bottle was tested for only 202 days, since the sample was 

 exhausted at this time. This sample was not transferred to the 

 pool. 



Each of the three samples was tested upon the same days and 

 in the same manner. The water in the two flower pots was thoroly 

 agitated by giving it a circular motion. With a sterile glass rod 

 the sides and bottom of the flower pot were scraped so as to loosen 

 adhering sediment that might contain tubercle bacilli. The water 

 in the bottle containing the bovine tubercle bacilli was thoroly 

 shaken before the test sample was taken. It was not convenient 

 nor thought necessary to rub the inside of this bottle, as was done 

 in the case of the flower pots. With a sterile Pasteur bulb pipette 

 from each of the three containers approximately 5 cc. of this in- 

 fected water were removed and placed in a sterile centrifuge tube 

 and centrifuged for thirty minutes at a high speed. Four cc. of 

 the supernatant liquid were removed and discarded. One or two 

 drops of the sediment were placed upon a glass slide and a micro- 

 scopic preparation made and stained for tubercle bacilli. The 

 remaining sediment and liquid in the centrifuge tubes, about i cc. 

 in quantity, was thoroly mixed and injected subcutaneously into a 

 guinea pig. The same precautions were taken with these tests as 

 in the case of the tuberculous manure to guard against mistaking 

 tubercles produced by dead tubercle bacilli for those produced by 

 living, virulent ones. Microscopic preparations and cultures were 

 always prepared to test the diseased tissues of the infected guinea 

 pig for tubercle bacilli. In a part of the cases, however, the tuber- 

 culosis was so extensive that it was not thought necessary to make 

 inoculations of diseased tissue into another guinea pig. 



RESULTS OF ^^ e resil l ts f the tests to determine the length of 

 THE TESTS time tubercle bacilli live in water are recorded in 

 Tables 22, 23 and 24. With the first four tests 

 of each of the three samples containing tubercle bacilli the guinea 

 pigs became extensively and severely tuberculous, as will be shown 

 by an examination of the three tables. Further tests of the dis- 

 eased tissue from all these guinea pigs showed the tubercle bacilli 

 to be living- and virulent. The fifth test, made after 126 days, 

 showed at least an apparent weakening of the virulence of these 

 germs. 



