464 



HOWARD B. NEWCOMBE 



From the interpretation here adopted, it follows that a sample from a culture 

 will contain some mutant individuals which are not yet phenotypically re- 

 sistant. When such a sample is spread on agar and allowed to grow (as in 

 method 4) these hidden mutants develop phenotypic resistance during the 

 first few divisions and thus give rise to resistant microcolonies. Since a mutant 

 clone may contain a number of these hidden mutants, and since these are dis- 

 persed over the surface of the agar, a number of resistant microcolonies can 



GROWTH 



Fig. 1.— Numbers of mutations to phage (Tl) resistance arising during bacterial multiplication, as 



estimated from the numbers of resistant microcolonies present after varying periods of growth. 



result from a single mutation prior to plating. This is of course not true for 

 mutations occurring after plating, because the products of these are confined 

 to one locality. 



Thus the rate of appearance of resistant microcolonies during the early 

 divisions would be expected to be high, approaching the true mutation rate; 

 and the rate of appearance during later divisions would be expected to be low, 

 approaching 2 n times the true mutation rate. 



Evidence has been obtained on this point, using method 4 and plotting 

 the numbers of resistant clones arising during varying periods of incubation 

 (R 2 — Ri, where time 1 is the time of plating) against growth in terms of factor 

 increase in the number of bacteria (N 2 /Ni). The absolute increase in resistant 

 clones with growth depends, of course, upon the original number of bacteria 

 plated, and where data are obtained from a number of separate experiments 

 they are comparable only if expressed in terms of the number of individuals in 



66 



