M. DELBRUCK 



657 



These predictions are borne out by the experimental results. Fig. 7 

 shows three such growth curves where nearly equivalent amounts of phage 

 and bacteria were mixed at time zero. The diminished yield is very 

 pronounced and the secondary step is discernible in two sets of observational 

 points. The condition of single infection of all bacteria at zero time can of 

 course be realized only approximately. Even if exactly equivalent amounts 



200 

 150 

 100 



50 



10 



pTb" 



S X 1.4 

 \ ^ 2.6 



CO 

 Hi 



3 



cleoring 

 I — "—I 



10 



20 TIME 30 IN MINUTES 50 



Fig. 7. Growth of phage if equivalent high concentrations (-^10 ^/cc.) of phage and 

 bacteria are mi.xed at time zero. Direct plot. 



Besides the experimental points from three growth curves three theoretical curves 

 are drawn in the figure. These are 



1. A one step growth curve with B in excess, taken from Fig. 4. 



2. A calculated growth curve, assuming inactivation of the liberated phage on 

 bacteria not yet lysed. 



3. Same, but assuming that the adsorbing power of the bacterial constituents respon- 

 sible for it is unimpaired till the completion of phage liberation and then vanishes 

 abruptly. 



The time interval from the beginning of clearing to its completion is indicated. It 

 falls well on the ascending part of the one step growth curve. In the one-to-one growth 

 curves this ascending part is soon counter-balanced by the multiple adsorption loss, 

 so that clearing seems to occur during a phase of little phage liberation. 



were mixed, the phage would not infect all the bacteria, but distribute 

 themselves according to the probability formulas derived by Poisson. If 

 P/B = n there will be a fraction e~" of the bacteria uninfected. On the 

 other hand our phage assays, though fully reliable as far as relative values 

 go, are not as certain with respect to absolute value, because of the difficulty 

 of obtaining an accurate determination of the efficiency of plating {cf. 

 Ellis and Delbruck (1)). 



It is not possible to make a complete quantitative prediction of the growth curve 

 because it is not known in detail how the adsorption en masse of phage to a bacterium 



71 



