M, DELBRUCK 655 



(Fig. 4) permits a closer analysis. Since the liberation of phage from the 

 individual bacterium probably occurs quite suddenly when the bacterium is 

 lysed our result means that the infected bacteria represent a mixture of 

 groups with latent periods ranging between 17 and 33 minutes and that 

 there is a uniform distribution of bacteria over this whole range of latent 

 periods. 



The question arises as to what causes a bacterium to have a shorter or 

 longer latent period. Several hypotheses might be suggested, either by 

 ascribing the cause to statistical fluctuations of reactions involving a small 

 number of particles (10), or by connecting it in one way or another with the 

 bacterial cycle. The latter view seems to the author the more likely one 

 but since it has not yet been worked out, further discussion will be deferred. 



Multiple Infection 



It was reported by Ellis and Delbriick (1) that if a bacterial suspension is 

 infected with an excess of phage no changes occur in the latent period or in 

 the burst size. At that time no phage concentrates were available and the 

 maximum ratio of phage to bacteria attained in that work was only four 

 to one. 



We have repeated this work with our new strains and with the concen- 

 trates and have been able to work with much higher ratios of phage/bac- 

 teria, up to 700 to 1. 



Fig. 6 shows some of the results obtained with the high ratios. Samples 

 were assayed every 3 or 5 minutes. Since the assays here require several 

 dilution steps these growth curves are less accurate both with respect to 

 assay values and with respect to timing. 



The results show an initial decrease in plaque count because many phage 

 particles are bound to one bacterium which then gives only one plaque. 

 For instance, starting with 10^" phage/cc. and a hundred times less bacteria, 

 one finds initially 10^° plaques/cc. After 10 minutes only 5 per cent of the 

 phage will be left free; these will give 5 X 10^ plaques/cc. In addition the 

 10^ bacteria/cc, each having adsorbed on the average 95 phage particles, 

 will give 10^ plaques/cc, bringing the total plaques to 6 X lOVcc. If the 

 initial ratio phage/bacteria is greater than a certain critical value the bac- 

 teria show saturation. This saturation value depends on the physiological 

 state of the bacteria. For instance, for rapidly growing bacteria, if the 

 ratio is 500, the plaque count decreases only by a factor two. The satura- 

 tion value is therefore 250. On the other hand, for 24 hour aerated bac- 

 teria, the saturation value is only about 20. (See Table III, last column.) 



We have indicated in the figure the time during which clearing of the mix- 



69 



