Rate of Change of Hereditary Factors. ii 



caused by the sorting out of old mutant factors into new com- 

 binations. There is, to be sure, enough work to show that the 

 real mutations are "rare" — whatever that term may mean; but, 

 so far as an approximate quantitative determination of the rate 

 of factor change is concerned, it is not possible, from the published 

 work, to determine even its general order of magnitude. Some 

 special scheme of crossing is required for this purpose. 



In the present series of experiments with Drosophila, the 

 X chromosome was chosen as the most convenient one for the 

 detection of mutation, since every hereditary factor in either of 

 the X chromosomes of the female fly stand revealed in the charac- 

 ters of one half of her male offspring, no matter what their father 

 was. Thus, if the female has a new mutated factor in one of 

 her X chromosomes, even though she does not usually show that 

 factor herself, and even though her mate does not contain it, 

 nevertheless one half of her sons are bound to show it and the 

 mutation will thus be recognized. There is reason to believe that 

 by far the commonest type of mutation is that which gives rise 

 to a lethal factor — which kills the organisms containing it — and 

 such lethal factors, also, in the X chromosome of a female, would 

 be revealed; in this case, by the fact that half the male offspring, 

 receiving it, would die before hatching. There would thus be half 

 as many sons hatched as daughters, giving a sex ratio of 2 9:1c 71 , 

 instead of i 9 : i cf , the usual ratio. In the first set of experi- 

 ments these lethal factors were looked for primarily, by making a 

 count of the sex ratios. 



As a preliminary measure, about 90 females were bred, and 

 the sex ratios of their progeny counted. Those families which 

 gave a lethal, i.e., 2:1, sex ratio (there were three of these) were 

 then discarded, since there was no way of knowing, in this pre- 

 liminary cross, whether these lethals had just arisen by mutation 

 or were of ancient origin. Females from the normal families 

 (with 1 : 1 sex ratio) were bred, however, since any lethal later dis- 

 covered in the descendants of these flies must be due to a really 

 new mutation, inasmuch as the ancestors had been certified as 

 normal. It was necessary, moreover, in selecting females for 

 breeding the next generation, not to breed many females from the 

 same family, but to choose them from as many separate families 



