GENES—STURTEVANT 
kinds of evidence that these plants are 
ultimately of hybrid origin, the two 
parents belonging, respectively, to the 
group of Old World cultivated types 
(the only forms known in the Old 
World before the discovery of Amer- 
ica), and the New World wild forms 
(of no economic importance).? Each 
of the two parental forms has 13 pairs 
of chromosomes; in the artificially pro- 
duced hybrid the maternal and pa- 
ABCDEF GHI JKLM 
ABCDEFGHI J KLM 
OLD WORLD 
299 
are the different species of Drosophila; 
in any case each of them has the full 
set of genes necessary to produce a 
cotton plant. When first produced, 
therefore, the form with 26 pairs of 
chromosomes had these genes present 
twice. In general in such a hybrid 
one of the duplicating pairs of genes is 
enough to produce a given character, 
and the other is likely to be lost by ac- 
cident. What has happened, then, ap- 
NOPQRSTUVWXYZ 
NOPQRSTUVWXYZ 
NEW WORLD WILD 
ABCDEFGHI J KLMNOPORSTUVWXYZ 
ABCDEFGHI J KLMNOPORSTUVWXYZ 
NEW WORLD CULTIVATED 
Ficure 2.—Chromosome constitution in the cotton plant. 
ternal chromosomes fail to pair, and 
the plant is sterile. Accidental dou- 
bling of this complement of 26 chromo- 
somes has given rise to the fertile New 
World cultivated types. 
The two parental types may be sup- 
posed to be related in the same way as 
8 It is not clear where or when these forms 
crossed to produce the New World cultivated 
types. These questions are of great interest, 
but cannot be discussed adequately here. 
pears to be that one set of genes nec- 
essary for the petal spot has been lost 
in some of the New World cultivated 
types, whereas in others this set has 
been retained, but the corresponding 
set from the other original parent has 
been lost. On this basis, then, it may 
be supposed that the differences in 
genetic determination are only ap- 
parent—the genes concerned are really 
the same, and are descended from 
