ACTIVE FROM INACTIVATED BACTERIOPHAGE 99 



that the "factor" has the same host specificity as the phage in whose lysate 

 it is found. 



The last point, particularly, was considered crucial in showing that produc- 

 tion of active phage from an inactive particle was actually due to infection of 

 the same bacterial cell with other particles, either of the same or of a different 

 but genetically related phage, active or inactive. Further confirmation came 

 from the experiments discussed below, which showed that in a mixture of 

 inactive phage and bacteria the number of bacteria yielding active phage is 

 never greater than the number of bacteria that adsorb two or more inactive 

 particles, and in some cases actually equals it. The same holds true for re- 

 activation of an inactive phage, for instance T2, by a related one, for in- 

 stance T4; the number of bacteria liberating active T2, is lower than — -or equal 

 to — the number of bacteria receiving at least one particle of each phage. The 

 analysis of cross-reactivation between these different wild-type phages will 

 not be discussed further in this paper, but will form the subject of a future 

 publication. 2 



(g) All bacteria which, after infection with inactive particles, do not liberate 

 active phage also fail to lyse. This was proved by mixing bacteria and inactive 

 phage under conditions in which some reactivation occurs, plating a sample of 

 the mixture for plaque count, and another sample for direct microscopic ob- 

 servation of lysis on agar. The results of such experiments proved that the 

 fraction of bacteria that are lysed is the same as the fraction of bacteria that 

 liberate active phage. The other infected bacteria fail to grow and divide, and 

 can be seen still apparently unchanged 24 hours later. 



Reactivation and genetic transfer 



The limitation of cross-reactivation to the T-even phages immediately 

 brought out a similarity between this phenomenon and that of genetic transfer 

 described by Delbrijck and Bailey (1946). In the latter case, bacteria 

 simultaneously infected with the phages T2r^ and T4r — the r character being 

 the result of mutation from the wild-type, which can be designated as r+ — liber- 

 ate a mixture of particles of the four types, r2f+, T2r, T4r+, and T^,. among 

 which the second and third represent new types. These must owe their origin 

 to some sort of recombination involving the genetic determinants for the 

 alternative r+ and r phenotypes. Evidence for the discrete nature of these 

 determinants has since been reported by Hershey and Rotman (1948). 



We assumed then, as a working hypothesis for the analysis of the reacti- 

 vation phenomenon, that inactivation by ultraviolet light resulted from 

 "lethal mutations" in a number of discrete genetic determinants among in- 

 active particles in the same bacterium to reconstitute fully active particles. 

 This hypothesis can be formulated quantitatively in terms of measurable 



* Cross-reactivation between T-even phages has the limitation that the individual phages are 

 distinguishable only by test of differential properties such as ability to grow on different hosts and 

 rate of inactivation by different antisera. Cross-reactivation can only be defined as the produc- 

 tion, upon mixed infection, of active particles having the distinctive properties of an inactive 

 parent particle. 



253 



