DELAYED EXPRESSION OF MUTATIONS 475 



and it is therefore tentatively suggested that the two delays are not of the same 

 origin. 



It should also be noted that a delay in phenotypic expression has been ob- 

 served in Paramecium following hybridization (Kimball 1937; Sonneborn 

 1943; Sonneborn and Lynch 1934) and has been termed "cytoplasmic lag." 

 This phenomenon is at least superficially similar to that described in the pres- 

 ent work, in that "phenotypic expression . . . commonly appears first at 

 slightly different times in different lines of descent within the clone" (Sonne- 

 born 1947). The extent of the delay appears to vary in Paramecium with the 

 nature of the change, and in two instances of alteration in type there is no 

 detectable delay. (These are the alterations from mating type I to mating type 

 II in variety 1, and from mating type V to mating type VI in variety 3. For a 

 discussion of this, see Sonneborn 1947.) It will therefore be of interest to 

 determine whether a similar diversity in the delay exists in bacteria, since the 

 apparent differences in the mutations so far studied, all of which mutations 

 are to resistance to one or another of the phages, are relatively small and can 

 be explained as due to known differences in the growth rates of the mutants. 

 Until more information of this nature is available it is not profitable to specu- 

 late further about whether "cytoplasmic lag" in protozoa and "delayed pheno- 

 typic expression" in bacteria have a common basis. 



conclusions 



Evidence of a delay in the phenotypic expression of spontaneous bacterial 

 mutations to phage resistance, using E. coli, strain B/r, has been obtained. This 

 evidence comes from a comparison of the rate of appearance of phenotypically 

 resistant clones during bacterial multiplication with the rate of gene mutation, 

 the latter being estimated from the numbers of resistant individuals arising 

 during similar periods of growth. 



The rate of appearance of resistant clones per bacterium per division cycle 

 is approximately 0.5 X10~ 8 , whereas the numbers of resistant individuals cor- 

 respond to a much higher rate of mutation, a rate of approximately 3X10 -8 . 



Since it is known that the mutants do not divide more rapidly than the 

 parent strain, the excess number of individuals composing a mutant clone in- 

 dicates that the clone was formed sometime prior to its becoming phenotypi- 

 cally detectable. Thus there appears to be a delay between mutation and pheno- 

 typic expression. 



The possibilities of alternative interpretations of the data of previous 

 workers on the basis of a statistical bias in one of the methods, and of a change 

 in the mutation rate during the later part of growth, have been eliminated by 

 using methods which are not susceptible to these sources of error. 



A delay in phenotypic expression would be expected to result in a high rate 

 of appearance of resistant clones during the first few divisions on agar, declin- 

 ing to the value 0.5 X10 -8 with further growth. This has been observed, al- 

 though an additional phenomenon appears to be present which results in the 

 appearance of a higher-than-expected number of resistant clones during the 



77 



