MUTATION AS A CHEMICAL PROCESS 



211 



hibits protein synthesis, these mutations could be prevented, suggesting 

 that they occurred as a result of the metabolic turnover of macro- 

 molecules. Of course, it is possible that some mutations may still take 

 place in nonreplicating genes by, for example, exchange reactions; we 

 may suspect, however, that most spontaneous mutations are copy-errors 

 in the course of gene replication. 



Similar conclusions may be drawn from studies of induced mutation. 

 Viruses, for example, in the free state cannot be shown to mutate spon- 

 taneously at a detectable rate, and they mutate only rarely even after 

 treatment with radiation. But if they are allowed to infect a cell, they 

 mutate with a small chance during their multiplication and abundantly 

 if the infected cell is irradiated. Even irradiating the bacterial cell 

 alone before infection induces mutations in the untreated viruses that 

 later multiply in it. Perhaps this is due to recombination between virus 

 and host, but we may suspect that viruses need to duplicate in order 

 to mutate, and furthermore, that mutation may involve the incorpora- 

 tion of unusual precursors during DNA replication. In this connec- 



FIGURE 8.4. The dependence of mutation on the residual replication of DNA in 

 Salmonella fyphimurium (fry ) put in the presence of chloramphenicol but no 

 tryptophane where the viable cell number remains constant. The mutagen, 2-amino- 

 purine, was added at the beginning of the experiment and after 2 hours when the 

 DNA had almost ceased replication (after Nakada, Strelzoff, Rudner, and Ryan, 

 1960, Zf. Vererb., 91:210-213). 



