448 HOWARD B. NEWCOMBE 



Methods of Previous Workers 



The early experiments of Luria and Delbruck (1943) referred to above 

 will now be considered in detail. 



To find out the number of phage-resistant mutants in a phage sensitive 

 liquid culture, the whole culture if its population size is small, or a sample of it 

 if it is large, is spread on agar together with the particular phage under con- 

 sideration. On incubation, all bacteria that are sensitive to the phage are 

 lysed, leaving only the resistant mutants. Each such mutant will eventually 

 form a separate colony; and from the number of colonies the number of mu- 

 tants in the liquid culture, or the sample, can be estimated. 



By using this technique it is possible to determine the rate of mutation to 

 phage resistance. As stated, two methods have been devised by Luria and 

 Delbruck (1943). The experimental procedure in both of the methods is to 

 grow a series of similar liquid cultures from small inocula, and to determine 

 the numbers of resistant bacteria in each, as well as in the average population. 

 The two estimates are derived from these primary data, the first using the 

 proportion of cultures in which resistant mutants have appeared, and the sec- 

 ond using the average number of resistant mutants per culture. 



These methods are of course only strictly applicable where it is possible to 

 eliminate the original type without affecting the mutant type, and where 

 there are no selective differentials between mutant and original types. 



In their experiments these test cultures were started with small inocula (50 

 to 500 bacteria from a growing culture of E. coli, strain B) and were grown to 

 saturation either in broth or in synthetic medium. (The volume of the cultures 

 was 10 cc in some experiments and 0.2 cc in others, the final numbers of bac- 

 teria being of the order of 3X10 10 and 3X10 8 respectively.) At the end of 

 growth, samples of the cultures — or in some cases whole cultures — were tested 

 to determine the numbers of bacteria resistant to the phage 77. 



The cultures from which the inocula were taken contained between 1 and 

 1,000 resistant individuals per 10 8 total bacteria. Thus, the chance of intro- 

 ducing a resistant bacterium into the test cultures via the inoculum was small 

 and in the rare event of one being introduced the fact would be indicated by an 

 excessive proportion of resistant bacteria in the fully grown test culture. In 

 practice, any resistant bacteria found at the time of testing — that is, after the 

 cultures are fully grown — -will therefore be the mutant offspring of one of the 

 sensitive bacteria in the inoculum. 



An estimate of mutation rate per bacterium per division cycle can be ob- 

 tained if the numbers of mutations, and the average number of cell 

 divisions, occurring in a series of cultures are known. The latter may be cal- 

 culated from the final population in the series and the former from the propor- 

 tion of cultures containing no mutants. The greatest accuracy is obtained when 

 this proportion is neither too large nor too small, and the method cannot be 

 used if every culture contains a mutant. Since the proportion of cultures having 

 no mutants is a function of the number of cell devisions in a culture, it may 

 be adjusted by altering the volume of medium, 0.2 cc being the amount used 



5° 



