360 



F. W, STAHL 



where B, the zero-dose extrapolate of the exponential portion of the curve, 

 is a function of x and n. Thus, a "critical test" (Dulbecco, 1952a) of the 

 hypothesis consists of the following: (1) the slope of log w (the experimentally 

 measured survival of multiply infected cells) versus r should for large values 

 of r be constant, independent of x, and equal to the slope found with single 

 phage particles, and (2) the back extrapolate to zero dose of the linear 

 portion of the plot of log w versus r should for each value of x give the same 

 estimate of n. 



Multiplicity reactivation in Vi-phage II of Salmonella typhi (Bernstein, 

 1957) is in agreement with the hypothesis (Fig. 3). Significant departures 

 from the hypothesis are found for T4, however. 



2 3 4 5 6 

 UVdose (mio) 

 Fig, 4. Multiplicity of ultraviolet-irradiated T4 in Escherichia coli B. The data 

 (siunmarized qualitatively by the smooth curves) are in agreement with the idea that 

 60 % of the phage target is completely reactivated in multiple infection; the remaining 

 40 % behaves as three "Luria units." The figure is adapted from Harm (1956). 



c. A Modification of the Luria Hypothesis. For T4 the plot of log lo versus r 

 at high dose has a slope which is significantly less than the slope for the 

 survival of a single particle (Harm, 1956). Harm has pointed out that an 

 explanation of this departure from the Luria hypothesis can be obtained by 

 assuming that the radiosensitive material of T4 is divided into two "parts." 

 One part is reactivated in multiple infection with an efficiency so high that 

 we shall ignore (for the moment) any small influence it might have on the 

 curve; the other part is composed of a small number of units which behave 



