310 BACTERIOPHAGES 



lower limit below which the mutation would pass undetected. 

 As a result, the r mutations in T4, which can occur in very many 

 loci, actually tend to recur rather frequently in a few (Benzer, 

 1957). Like the facts mentioned above concerning h muta- 

 tions, this shows that stability at a locus depends on local details 

 of structure. The same conclusion is supported by the sugges- 

 tion that different mutations at the same locus can be associated 

 with different reversion frequencies at that locus (Benzer, 1957). 

 So far there is no evidence that stability at a given locu ^ is influ- 

 enced by genetic structure at distant points. Some early 

 indications of this (Hershey, 1946a) have not been borne out by 

 further study (N. Symonds, personal communication). 



Luria's (1951) demonstration of a clonal distribution of r 

 mutants among single cell yields of T2 provides the chief evi- 

 dence for a geometric mechanism (as opposed to an observed 

 linear rate) of phage growth. 



5. Mutagenic Agents 



There is as yet no indication that mutagenic agents produce 

 mutations in extracellular phage particles. In contrast there is 

 considerable evidence that these agents have a pronounced 

 mutagenic effect when applied to infected host cells. There was 

 much scepticism concerning early papers on mutagenesis be- 

 cause of the difficulty of ruling out possible selection by the muta- 

 genic agent. However, the demonstration of the phenomenon 

 with several different mutagens now seems convincing. 



Apparently the first experiments were performed by Latarjet 

 (1949), who produced h mutations in T2 by ultraviolet irradi- 

 ation of infected bacteria. The infected bacteria were plated on 

 B/2 to determine the number of infected cells that liberated h 

 mutants and on B to determine the number that liberated any 

 phage. The ratio between these two numbers was taken as a 

 measure of the frequency of occurrence of mutations. This 

 frequency for unirradiated controls was 15 X 10"''. Irradia- 

 tion during the first third of the latent period before much DNA 

 synthesis had occurred gave doubtful and inconsistent results. 



