MANNER OF PRODUCTION OF MUTATIONS 545 



to mutations by Novick and Szilard (1949), working on E. coli. It was 

 shown that the reparative effect for mutations closely paralleled that for 

 the killing action (inactivation) in all features studied, as though a com- 

 mon process or substance, of mutagenic potentialities and on which both 

 results depended, had been affected. Since the time required for the 

 possibility of repair to fade away extended over more than an hour (i.e., 

 since at an interval of an hour after application of the mutagenic light, 

 some repair could still be effected), it became evident that some of the 

 mutations had not yet been effectuated, or at least completed, before a 

 period of this order had elapsed. It is possible that some mutations may 

 occur much later, but, if so, these would no longer be susceptible to the 

 reparative influence. 



The reparative effect of long ultraviolet and/or visible light in the 

 production of mutations by mutagenic ultraviolet has been confirmed in 

 E. coli by Newcombe (1950) and has been extended to various other 

 organisms, including Paramecium (Kimball and Gaither, 1949), Micro- 

 coccus (Haas et al., 1950), Neurospora (Goodgal, 1950), and Drosophila 

 (Meyer, 1951; Altenburg and Altenburg, 1952). In all these organisms 

 the effects on cell survival paralleled in a general way those on mutagene- 

 sis, as though a stage antecedent to and common to both had been 

 affected. In some of the organisms the repairable portion p of the effects 

 of the mutagenic light was much higher than in the bacteria studied. 

 Thus, in Drosophila pole cells, it was possible to prevent about 75 per cent 

 of the total mutational effect and in Neurospora, 95 per cent. Of course, 

 this could be made evident only at lower doses, at which the frequency- 

 dosage curve was sharply rising. 



It is possible, as already noted on pp. 536 537, that in some of the 

 organisms studied even the mutagenic ultraviolet, over at least a part of 

 its range, exerts a significant amount of reparative effect in addition to its 

 directly mutagenic effect. It is true that, with the phage studied, repair 

 was found to be practically absent at wave lengths approaching the muta- 

 genic ones in shortness and that with E. coli the constancy of the effect- 

 dosage relation over a considerable range of dosages virtually precluded 

 the possibility of a reparative effect by ultraviolet of the wave length 

 employed; but, since different microorganisms have been discovered to 

 have different action spectra for repair, it may well be that in other forms 

 the repair spectrum does overlap that of mutagenesis. It would be diffi- 

 cult to test for this because of the superposition of the two effects. How- 

 ever, their simultaneous existence would help to explain why, in so much 

 of the material thus far studied, the ultraviolet mutation frequency 

 reaches a ceiling or a peak followed by a decline, where factors of irregular 

 illumination, sensitivity, and selection seem insufficient to furnish a com- 

 plete explanation of the extent of the drop in efficiency. In addition to 

 the possibility of repair by the mutagenic ultraviolet itself, there is the 



