474 HOWARD B. NEWCOMBE 



dividual may be obtained by comparing the estimates of mutation rate from 

 method 1, utilizing the number of resistant clones developing in liquid cultures, 

 with those from method 5, utilizing the number of resistant individuals in old 

 mutant clones. Table 9 may be used for this purpose, and the ratio rate (5)/ 

 rate (1) considered. If one omits the data on mutations to B/3, 4 and B/3, 4, 

 7, because of the bias resulting from the known low growth rate of these mu- 

 tants (Demerec and Fano 1945, table 7), the ratios in the cases of the remain- 

 ing mutations appear to be similar. A crude average of these is 6.4. This in- 

 dicates that there are approximately six or seven individuals in a mutant clone 

 at the time of appearance of the first resistant individual; that is, that the 

 delay between mutation and phenotypic expression in at least one member of 

 the resulting clone is of the order of 2.7 generations. 



The extent of the delay in individual lines of descent within a mutant clone 

 other than the first one to become resistant, cannot be determined directly. 

 However, the absence of any striking difference between the estimates of 

 mutation rate obtained by methods 2 and 5 suggests that the delay is not ex- 

 cessive; and the similar values, 2.8X10~ 8 and 3.4X10 -8 , obtained by method 

 3 with growth periods of six and eleven generations respectively, indicate that 

 no very large proportion of a mutant clone changes to phenotypic resistance 

 after six generations from the time of the mutation. 



The possibility that the delay in expression of irradiation-induced changes 

 is of the same origin as that observed in spontaneous mutations should be con- 

 sidered at this point, although a positive answer cannot be given, since strictly 

 comparable studies of the induced changes have not yet been carried out, and, 

 in particular, nothing is known of the numbers of resistant individuals develop- 

 ing in the late-appearing resistant clones resulting from irradiation. 



If irradiation results in the immediate induction of gene changes similar to 

 those occurring spontaneously, and the delay is one of phenotypic expression 

 similar to that in untreated material, then the numbers of resistant individuals 

 in the late-appearing resistant clones should rise rapidly after the clones first 

 become detectable, because of susceptible members becoming phenotypically 

 resistant, and should eventually approach the numbers in the resistant clones 

 that appeared soon after treatment. Variation in the time of occurrence of 

 phenotypic expression in different lines of the same mutant clone would also 

 be expected in irradiated material. 



Until information of this nature is obtained it is not possible to say with 

 certainty that the observed delays in appearance of induced and spontaneous 

 mutations are of the same origin. In the absence of more detailed information, 

 one can only speculate on the basis of the variation between clones. Since in 

 untreated material there is no striking difference between the apparent muta- 

 tion rates as determined over periods of six and eleven generations (method 3), 

 it would seem that six generations is adequate both for phenotypic expression 

 in most of the mutants within a clone and also for expression in at least one 

 individual in most mutant clones. In irradiated material an appreciable pro- 

 portion of the resistant clones do not appear until after the sixth generation; 



76 



