BACTERIOLOGY OF MILK 223 



on the contrary, tubercle bacilli may be missed. Should 

 organisms morphologically resembling the tubercle bacillus 

 be found in a sample from a herd, inspection of the beasts 

 will often reveal an animal showing signs of the disease. 

 The milk from such must be proscribed, pending the 

 result of an animal experiment: Two guinea-pigs are 

 inoculated, one subcutaneously, one intraperitoneally, 

 each with 1 to 2 c.c. of the sediment. 



Although probably mistakes do not often occur, the 

 guinea-pig test is not infallible. The experimenter may 

 be misled by the lesions produced by Timothy-grass 

 bacilli, Rabino witch's bacillus, the ' Mistbacillus ' or 

 Johne's bacillus. Conversely, error may occur in the 

 opposite direction, for negative results may be obtained 

 with guinea-pigs inoculated with milk from tuberculous 

 udders. The test usually occupies four or six weeks. 

 In spite of these disadvantages, it is the most reliable 

 method at present known for detecting tubercle bacilli. 



Examination for the Diphtheria Bacillus. A number 

 of serum-tubes are inoculated. If an organism resembling 

 the Klebs-Loffler bacillus be isolated, it must be sub- 

 mitted to the test of inoculation, since bacilli are not 

 infrequently present in milk and milk products which, 

 though resembling the Klebs-Loffler bacillus morpho- 

 logically and culturally, are non- virulent. 



Examination for the Typhoid Bacillus. The milk can 

 be examined by one of the methods described under 

 ' The Examination of Water.' 



Enumeration of Organisms. This is not generally per- 

 formed as a routine test, and there are various objections 

 to attempts to use it as an indication of purity. Should, 

 however, a count be desired, plates must be made on the 

 decimal system to show a range of organisms from 20,000 

 to 20,000,000 organisms per cubic centimetre. Dilutions 

 of from 1 in 1,000 to 1 in 100,000 should be made. Eastes 

 recommends distilled water agar. 



Counts made directly on microscopical preparations o 

 milk are higher than those obtained by plating (Brew). 



Estimation of B. Coli. Quantities of 10 and 1 c.c. of 

 milk are inoculated into MacConkey bile-salt lactose 

 peptone medium tubes, and decimal dilutions containing 

 from 0-1 to O'OOOOOl c.c. of milk are inoculated into tubes 

 of the same medium, using the same sterile 1 c.c. pipette, 



