D. E. Lea and C. A. Coulson 271 



is the probability that a culture in which exactly j mutations have occurred shall contain 

 r mutants. In practice when comparing the theoretical and experimental distributions the 

 distributions will always be grouped. For economy of space, therefore, we do not publish 

 the full table of C jr but give in Table 1 grouped values for r = l, 2, 3, 4, 5-8, 9-16, 

 17-32, 33-64 and >64. 



Table 1. C Ur 



Cj T is the probability that a culture in which exactly j mutations have occurred shall have r mutants (r^j). 

 For values of r greater than 2, the values of Cj r have been grouped. Thus the numbers in the column headed 



r=32 



' 17-32' are values of S C.- r , See also the Appendix for certain other values of C j>r . 



r=17 ' 



p r is the probability that a culture of such a size that the mean number of mutations is 



r / m?\ 



\ shall contain r mutants. p = e~ m and for r^l, p r = 2 C jr \e~ m ^-A. For any given 



j = \ \ J • I 

 alue of m, p r can be calculated with the aid of the table of C jr , for the same groupings 



iir 



of r. The calculation is facilitated if a table of Poisson coefficients e~ m -rr is available 



(Molina, 1942). In Table 2 values of p r are given for a number of values of m from 0-05 

 to 15. 



The figures in Tables 1, 2 and 4 are liable to occasional rounding errors of one unit in 

 the last decimal place. 



Journ. of Genetics 49 19 



3* 



