DISTRIBUTION OF BACTERIA. 149 



the fat. It may be diluted one hundred times with steril- 

 ized water; when the number of bacteria is very great a 

 second dilution may be required. Estimations based upon 

 such high dilutions can only be approximate. The quan- 

 tity taken for examination may be o.i to i c.c. Plates 

 should be made immediately upon collection of the sample. 

 If the milk stands for a few hours at the room temperature 

 in the laboratory, the number of bacteria will become 

 enormously increased. 



The detection of a particular kind of pathogenic bacterfa 

 in milk or butter involves very great difficulties. Staining 

 of bacteria in milk may be done by the usual methods, but 

 the results are rendered unsatisfactory by the oil in the milk. 

 The demonstration of tubercle bacilli by staining methods is 

 likely to involve many difficulties. In this connection it is 

 necessary to remember the group of bacilli which resemble 

 the tubercle bacillus in resisting decolor ization with acids 

 after staining. (See p. 44.) The procedure of injecting 

 milk into guinea-pigs has been resorted to largely, but the 

 results are only obtained after the lapse of weeks, when the 

 development of tuberculosis in the guinea-pigs would indi- 

 cate that the milk was tuberculous, provided that control 

 guinea-pigs remained healthy. Furthermore, the other acid- 

 proof bacilli which may occur in milk or butter are capable 

 of producing nodules resembling tubercles. 1 (See Bacillus 

 tuberculosis, Part IV.) The most satisfactory plan will be 

 to apply the tuberculin test to the cow from which the milk 

 was derived. 



Among the other articles of food, those are to be most 

 carefullv scrutinized which are to be eaten after little or no 

 cooking, particularly salads, green vegetables, fruits, and 

 the like. Under exceptional circumstances they may be- 

 come agents for conveying infectious diseases. Conn 



1 Rabinowitsch, Zeitschrift f. Hygiene, Bd. XXXVII. , p. 439. 



