Vol. 6, 1920 
GENETICS: A. WEINSTEIN 
627 
magenta, and forked in the other. In the offspring of this female (number 
1309, table AD), which had been mated by yellow males, it was noted that 
all the non-vesiculated sons were crossveinless. This indicated that the 
locus of the new factor was in the X chromosome very near the vesiculated 
locus, and that the mutation had occurred in the yellow-bearing chromo- 
some or in one of its ancestors. This chromosome had come from yellow 
stock, and an examination of the stock showed that a large proportion 
of the yellow flies were crossveinless. The success of the mutation in 
establishing itself in the crowded conditions of a stock bottle indicated a 
FIG. 2 
(a) Wild-type wing of D. melanogaster (the wild-type venation of D, virilis is identi- 
cal with this); {b) crossveinless wing of melanogaster; (c) crossveinless wing of virilis. 
good viability, and subsequent experiments have shown that the viability 
is comparable to that of wild-type flies. The excellent viability and 
the clear-cut nature of the crossveinless character will make it very useful 
in determining linkage relations in D. virilis. 
Crosses involving yellow, crossveinless, and vesiculated (table 1) show 
that the crossveinless locus lies between the other two, very near the 
vesiculated locus. The yellow crossveinless distance in these experi- 
ments was 19.23 units and the crossveinless vesiculated distance 1.15 
units. The inclusion of other data (tables 2 and 3) lowers the yellow 
crossveinless distance to 17.6 units. 
The determination of the crossveinless vesiculated distance may be 
subject to an error due to the fact that vesiculated is not a perfect diag- 
nostic character. Vesiculated flies show the character sometimes in both 
wings, more often in only one, and occasionally (as we should therefore 
expect) in neither wing. This last point is proved by the fact that matings 
