444 H. J. MULLER 
factor could not be secured, on account of the lethal action of 
the deficiency, the dilution of eye color which it produces when 
in combination with the other allelomorphs of W shows that in 
its color effect it is an ‘ultra-white.’ Each of the mutant allelo- 
morphs of W arose by a single mutation from the normal gene, 
excepting eosin, which arose by a mutation of the gene white, 
and which is therefore removed from normal by two mutations. 
All of the allelomorphs affect the same character, eye color, and 
together they form a graded series.1 And, in addition to the 
mutations originally observed, certain of the factors—w and 
possibly we—have been found to arise more than once: white 
having arisen several times by ‘reverse mutation’ from eosin 
and the normal red eye color once reappearing by ‘reverse muta- 
tion’ from white (Morgan and Bridges, 719). The locus W 
therefore represents the nearest approach yet found in the 
fruit-fly to the supposititious condition of factor fluctuation 
which most selectionists have postulated, and the findings con- 
cerning it have in fact already been made use of by Jennings 
C17 a, b) as an argument in favor of such views. Four more 
mutations, one of which probably, and three of which certainly, 
belong in the same series may now be reported; one of the latter, 
ivory, was found by Sturtevant, whose own account of it he has 
kindly allowed to be incorporated in the present paper (section 
II); the other three mutations were found by the author. It 
will be of interest to examine these mutations with reference 
to the question of factor fluctuation raised above, and also to 
consider the case as a whole in its present relation to the prob- 
lem of mutation in general. The data concerning each of the 
new mutants will be given first. 
1JTt had been thought (and mentioned in the literature) that these mutant 
allelomorphs also affected body color, inasmuch as flies containing them appear 
lighter than red-eyed flies after being killed and ‘extracted’ for several days in 
50 per cent alcohol. I have found, however, that decapitated red- and white- 
eyed flies show no such difference after treatment with alcohol. The effect is 
obviously due to the red-eye color of the normal flies becoming partly dissolved 
by the alcohol and distributed through the body of the fly; flies with lighter eyes 
have less color to be thus distributed. 
