THE MANNER OF OCCURRENCE OF MUTATION 457 
somehow interacted upon the eye color in just such a way as to 
exactly counteract, in the case of six separate mutants of W, 
the eye-color difference that would be caused by the difference 
in quantity of W carried by the male and female. This would 
indeed be a remarkable series of circumstances, especially since 
it would have to be extended so as to apply not only to these 
mutants of W, but to practically all the mutant and normal 
genes known in other loci of the X chromosome, for in all these 
cases the male shows the same intensity of character as the 
corresponding homozygous female. As a matter of fact, when 
we come to examine those cases in which such an interaction of 
sex with other characters could be demonstrated—namely, the 
cases of non-sex-linked mutant characters—we find any kind of - 
interaction of sex with other characters to be relatively rare, 
let alone that sort of very nicely adjusted interaction that would 
be required for the present purpose of just counterbalancing 
the difference in quantity of the gene in the two sexes. 
It is accordingly evident that the similarity in eye color of 
males and homozygous females carrying some mutant of W is 
due to the fact that a difference of as much as 50 per cent in the 
gross amount of a certain kind of gene received by the fertilized 
egg does not ordinarily affect the manifestation of the gene. 
Hence it follows also that the lighter the color of, for example, 
an écru-white heterozygote, as compared with homozygous 
écru, cannot be due to the reception by the former of a smaller 
total quantity of eye-color-producing gene material, but must 
be due to the gene for white being qualitatively different from 
that for écru, and exerting a positive influence to make the eye 
color lighter. Further evidence along the same line is furnished 
by the fact that not all the colors given by the W series are 
exactly alike in kind: although on the whole they form a graded 
series, still buff, for example (in spite of its name), gives a dis- 
tinetly clearer yellow color than either tinge, which is just above 
it in the scale, or écru, which is just below it. Moreover, the 
eosin and ivory colors are both affected by sex, whereas none of 
the other factors, either above or below them in the scale, are so 
influenced. 
