ANDRE LWOFF AND ANTOINETTE GUTMANN 



In contrast, lysozyme lysis does liberate bacteriophages adsorbed to killed 

 bacteria (A. Pirie, 1940). The bacteriophage virus does not transform itself into 

 probacteriophage except in living bacteria. 



The importance of these facts stimulated us to verify the action of lysozyme 

 lysis. Washed bacteria are placed into microdrops. In the course of exponential 

 multiplication, one injects into the microcultures a solution of lysozyme of such 

 concentration that its final dilution is 1 to 50,000. The bacteria are lysed in 

 several minutes. In different experiments 12, 34, 80, and 208 bacteria were thus 

 lysed. All of the liquid was sampled: no bacteriophages. The presence of 

 lysozyme at the concentration of 1 to 50,000 in a microdrop does not interfere 

 with the assay because the medium is considerably diluted. These observations 

 thus extend to exponentially growing bacteria the results of E. and E. Wollman 

 and of A. Gratia on B. megaterium and they confirm the fundamental observation 

 of F. Burnet and M. McKie (1929b). 



VI. How Do Lysogenic Bacteria Liberate Bacteriophages? 



We are thus in possession of the following facts: 1. In the lysogenic strain of 

 B. megaterium 899, every bacterium isolated in the micromanipulator and 

 washed gives rise to a lysogenic colony. 2. The faculty of producing bacterio- 

 phages perpetuates itself in the absence of exogenous bacteriophages, that is to 

 say by an endomicrobial route. 3. Bacteria can multiply without liberating 

 bacteriophages. 4. Lysozyme lysis does not liberate bacteriophages. 



Nevertheless, a filtrate of a culture of lysogenic B. megaterium, indeed the 

 filtrate of every lysogenic culture, always contains some bacteriophages. We 

 have verified the fact that this is also true for dense cultures in microdrops: one 

 always finds bacteriophages in the liquid of a small drop of peptone water which 

 contains several thousand bacteria. One can thus entertain the hypothesis that 

 only a small percentage of the bacteria produce bacteriophages. We were 

 encouraged to persevere in the investigation of the mode of liberation of bacterio- 

 phages by the observation of bacterial "ghosts." The formation of the ghosts 

 was observed in the microdrops. It is the result of a relatively slow degeneration 

 of the bacterium which can take several minutes or several seconds. The ghosts 

 persist for several hours. We thought that we had observed bacteriophage lysis. 

 But assay of the medium after the slow lysis did not reveal any bacteriophages. 

 Nevertheless, we continued to follow the development of the microclones, 

 assaying the medium from time to time. We give here the results of a first series 

 of experiments published previously in a preliminary note (A. Lwoff and A. 

 Gutmann, 1949b). 



A. A filament of free bacteria is washed and brought into one drop of medium 

 S. The following table gives the time in minutes (m), and the number of bacteria 

 (b): 



m h m h 



319 



