334 F. JACOB AND E, L. WOLLMAN 



of phenomenon forms to a great extent the basis of the differences between 

 various strains of the same species as classified by phage typing. 



C. Lysogenic Conversions 



Other alterations in the characters of the host as a result of lysogenization 

 appear to be so entangled with the lysogenic character than they can be 

 ascribed to the very presence of the prophage. The most striking case is the 

 production of toxin by Corynehacterium di'phtheriae. It was observed by 

 Freeman (1951) tha,t most of the toxinogenic strains are lysogenic and release 

 phage particles which are active on other nontoxinogenic, nonlysogenic 

 strains. The lysogenic clones which can be isolated by infection are all toxi- 

 nogenic and the toxinogenic character can be passed from strain to strain by 

 lysogenization (Groman, 1953). The toxinogenic character disappears when 

 lysogeny is lost. A complete correlation is observed between lysogeny and 

 toxinogeny (Groman, 1955). However, such a property appears to be re- 

 stricted to certain strains of temperate phages of C. diphtheriae (Barksdale, 

 1955; Groman and Eaton, 1955). Experiments by Groman and Eaton also 

 suggest that the ability to confer toxinogeny segregates in crosses between 

 related phages. 



The exact nature of the relationship between lysogeny and toxinogeny is 

 not yet clear. Phage and toxin are antigenically unrelated. It seems milikely 

 that toxinogenic bacteria result from the selection by the phage of pre- 

 existing toxigenic mutants (Groman, 1953). Toxin production is not corre- 

 lated with vegetative multiplication of phage (Barksdale and Pappenheimer, 

 1954). The favored hypothesis is that the very presence of some types of 

 prophages confers on certain bacteria the ability to produce a specific protein. 



In B. megatherium, a case has been reported by lonesco (1953) in which 

 the presence of a given prophage modifies the colonial morphology. The 

 colonies revert to "normal" after the loss of the prophage. 



In Salmonella, the presence of certain somatic antigens appears to be 

 controlled by the presence of certain prophages, as shown by Iseki and 

 Sakai (1953). Strains of subgroup Ej possess antigens III, XV. They are 

 lysogenic and release phage e. Strains of subgroup Eg possess antigens III, X. 

 They are sensitive to phage e. Upon infection of bacteria Eg with phage e, 

 the lysogenic derivatives E2(e) become antigenically III, XV. The presence 

 of prophage e appears therefore to be correlated with the formation of the 

 somatic antigen XV. Likewise, in Salmonella group B, the presence of pro- 

 phage "iota" appears to control the formation of somatic antigen I (Iseki 

 and Kashiwagi, 1955). 



These various observations indicate clearly that a prophage may eventually 

 interfere with the biochemical activities of the host cell. Such alterations of 

 the host physiology resulting from the presence of a prophage might possibly 



