232 THE BACTERIOPHAGE AND ITS BEHAVIOR 



the culture in a liquid medium hardly differs from that of ultrapure 

 cultures of bacteria of the same species. When dealing with bacteria 

 belonging to a homogeneous species it is easy to prove ultrapurity, for 

 it is only necessary to filter a bouillon culture of the bacterial strain in 

 question and to combine this filtrate with a normal organism of the 

 same bacterial species. If bacteriophagy takes place the culture is an 

 artificially mLxed one; if it does not, the strain is certainly ultrapure. 

 But with regard to bacterial strains belonging to heterogeneous species 

 the proof may be far more difficult, although we will see that it is still 

 possible. 



Gildemeister seems to have been the first to notice that certain strains 

 of intestinal bacteria might give, on agar, abnormal colonies which he 

 termed "Flattenformen," — of variable aspect. He noticed that 

 subcultures made by the isolated colony method from such 'Tlatten- 

 formen" gave colonies of different appearances. He concluded that 

 certain strains of intestinal baciUi were subject to mutations, without, 

 however, recognizing that these mutations took place through the 

 influence of a principle separable from the bacterium. 



Bail^^ has succeeded in isolating bacteriophages from 3 strains of 

 Flexner dysentery baciUi. 



Moreover I have observed that in certain cases of cystitis the blad- 

 der contains, not a pure culture of B. coli, but a mixed culture com- 

 posed of resistant B. coli and a bacteriophage. Platings of the urine 

 upon agar yield ultrapure colonies and mixed colonies from which it is 

 possible to isolate a bacteriophage.^21 



Otto and Munter,'*^^ Weinberg and Aznar®^^ and Seiffert"^ have also 

 isolated bacteriophages from different strains of intestinal bacilli, but 

 these authors have attempted to generalize and have affirmed that all 

 bacterial strains contain a dissolving principle. Jotten^^^ has expressed 

 the same opinion although he admits that "the experiment is not always 

 successful." As for Gildemeister and Herzberg^^^ ^j^gy admit that many 

 strains appear to be free of bacteriophage but they suggest the hy- 

 pothesis that when filtrates of old cultures appear to be free of all "lyso- 

 genic" action this must be because they are not acting upon a susceptible 

 bacterium. If all bacterial species were heterogeneous with regard to 

 the bacteriophage such an hypothesis would be open to discussion but 

 the existence of homogeneous species obviously renders it inadmissible. 

 If we take a secondary culture or a mixed culture of dysentery bacilli 

 and filter it through a candle and then combine the filtrate with a normal 

 strain of any B. dysenteriae whatsoever, bacteriophagy takes place. 



