V. MOLECULAR MECHANISM OF MUTATIONS 263 



than 50% of the revertants gave standard phenotype while the rest gave 

 large fuzzy plaques on bacteria B. The important point is that each of 

 these four mutants does yield at least some induced revertants of 

 standard phenotype. When these revertants were backcrossed to standard 

 type phages the frequency of r mutants observed in the progeny was not 

 significantly higher than in the stocks used (usually smaller than 0.1%, 

 Freese et al., 1961b, E. B. Freese, 1962). As outlined before these results 

 make it likely that the revertants of standard phenotype arose by 

 genuine back mutations. The four AP mutants in Table V must then 

 have a G-C pair at their mutant site. 



The mutant AP 12, although highly inducible, has been omitted from 

 Table V because a functional test revealed that all induced revertants 

 produce r plaques on bacteria B and the same was observed with another 

 mutant of this genetic site. Revertants of AP 12, induced by the above 

 agents, may arise only or mainly by suppressor mutations. 



The data in Table V indicate that BU induces transitions from GC 

 into AT more frequently then in the opposite direction ; this has already 

 been explained earlier (Section III,B,1). In contrast, AP seems to induce 

 transitions in both directions with comparable frequency. Actually all 



G A 

 other transition inducing agents predominantlv induce changes -^ . 



C T 

 The observed directional effect of the base analogs complements the 

 earlier observation that AP is more efficient than BUdR in inducing the 

 reversion of BU-induced mutants (Freese, 1959b) ; the opposite state- 

 ment concerning AP mutants seems no longer justified. Similar direc- 

 tional effects for induced reversions have been observed in Salmonella 

 typhimurium (Margolin and Mukai, 1961) when the frequency of AP- 

 induced revertants was measured on plates with minimal and with 

 suboptimally enriched medium. Again, kinetic studies, on the time and 

 mode of appearance of bacterial revertants induced by BUdR or by 

 AP, have been used to identify the mutant base pair in a transition 

 mutant fRudner, 1961; Strelzoff, 1962). 



5. Specificity of Spontaneous Mutations 



The preceding analysis on the induction of mutations also provides 

 infoniiation about the specificity of spontaneous mutations. The result 

 will be summarized for the mutations of phage T4: 



1. About 80% of the spontaneous mutations are of the non-transition 

 type; the rest apparently arose by transitions (see Table III). 



2. Some of the spontaneous non-transition mutants, including some 

 hot spots, seem to cany a transversion (see Table IV). The other non- 



