INTRACELLULAK MULTIPLICATION OF BACTERIAL VIRUSES 243 



until lysis of aU the infected bacteria is certain to have occurred and the total 

 contents of each tube plated for assay of the number of infective progeny. 

 In the experiment of Ellis and Delbriick, 25 out of 40 tubes, or a fraction of 

 0.62, showed no plaques upon plating, i.e., did not contain any infected 

 bacteria at all. The remaining 15 tubes showed anywhere from 5 to 190 

 plaques, of which, according to (1), about 12 plates probably represent the 

 yield from a single infected bacteria. The burst sizes of individual infected 

 cells are thus seen to be subject to considerable variations. More extensive 

 experiments by Delbriick (1945a) showed these variations to be extremely 

 wide indeed, since rare bacterial cells were observed which Hberated more than 

 1000 progeny phages. An average burst size may also be calculated from the 

 single-burst data by dividing the total number of plaques on all plates by the 

 number of infected bacteria which have given rise to the entire progeny 

 brood. This ratio is usually found to be similar to the average burst size 

 estimated directly from a one-step growth experiment. 



The reasons underlying the very broad distribution of individual burst 

 sizes still remain obscure. Delbriick (1945a) noted that the variation in 

 length of individual bacteria of the infected culture was very much less than 

 the variation in number of progeny particles which these cells hberate. 

 Delbriick inferred, therefore, that variable ceU size was probably not the 

 factor involved in burst size variation. It seems possible that it is the complexity 

 and, in part, geometric nature of the process of intracellular phage growth 

 itself which is responsible for the wide fluctuations in phage yields of individ- 

 ual infected cells. For if the final number of mature progeny phage particles 

 liberated at the time of lysis depends on a number of consecutive as weU as 

 concurrent reaction steps, then relatively small perturbations in any one 

 of these reactions can easily lead to a very considerable diminution of the 

 maximum number of phages of whose synthesis the cell is capable under very 

 optimal conditions. 



C. Lysis and Lysis Inhibition 



The kinetics of bacteriophage growi)h may be followed not only by assaying 

 the infective titer in one-step growth experiments, but also by observation of 

 the lysis of the infected cells, either through direct microscopic examination 

 of the bacterial cells or through macroscopic measurement of the turbidity of 

 the bacterial culture. In such experiments it is necessary, of course, that 

 nearly aU of the bacteria are infected at the outset of the experiment, i.e., 

 that the average number of phage particles adsorbed per bacterium, or the 

 multiplicity of infection, is greater than one. By following the time course of 

 lysis in a typical one-step growth experiment, it can be seen that the number 

 of intact bacteria visible under the microscope and the turbidity of the 

 infected culture begin to decrease precisely at the conclusion of the latent 



