446 H. J. MULLER 
Since males receive all sex-linked factors from their mother, 
this factor must have arisen by mutation in an ovarian cell of 
the mother of the original écru male (either before or after its 
fertilization, but probably before its cleavage). Since none of 
the brothers of this male were likewise écru the mutant gene 
could not have existed in many of the egg cells of the mother, 
hence the mutation could not have occurred in an early oogonial 
stage, but probably took place in a late oogonium or in an oocyte. 
B. Locus, and mode of reaction with other members of the W series 
Ecru males were then crossed to white-eyed females, and pro- 
duced daughters intermediate in color between écru and white. 
This showed that écru was either an allelomorph of white, or 
that, when present, it caused white to be partially dominant. 
To decide between these two unequal possibilities, the F, from 
the cross of écru by white were then bred together, and it was 
found that no red-eyed crossovers were produced; écru therefore 
lay in the same locus as white, or was completely linked to it; 
in other words, the two mutants were allelomorphs. 
Crosses of just the same type were performed with écru and 
eosin, and with precisely similar results. ‘The F; females were 
intermediate in color between écru females and eosin females, 
and in subsequent generations no crossing over took place be- 
tween the two factors. ‘There is consequently no question that 
wee is a member of the W series of allelomorphs, and that it 
behaves like the other mutants of the series in being recessive 
to normal (W), and in giving intermediates when crossed with 
white, eosin, and, presumably, with the other mutant members 
of the series. 
II. LVORY? 
A. Manner of origin 
In an experiment designed to fix an increased value of crossing 
over in a part of the second chromosome, brother-sister pair 
2 The results in section II were obtained by Dr. A. H. Sturtevant, and nearly 
all of this section was written by him. 
