ica and is of special interest in that it has recently been 
the subject of detailed cytogenetic investigation (105). 
The cytological evidence shows that the New World 
cottons are allopolyploids (amphidiploids); that is, a type 
of stable hybrid which arises through the doubling of 
chromosomes in the progeny of an interspecific cross. 
One of the parents of the American allopolyploid cottons 
was G. Raimondu Ulbrich of Peru, or a similar (perhaps 
ancestral) type, while the other was an Old World type 
similar to G. arboreum L. While the explanation offered 
by Hutchinson, Silow and Stephens (105), that the Old 
World cotton was carried across the Pacific by man, has 
been the subject of some controversy, we are here pri- 
marily concerned with the evidence that the American 
cultivated cottons arose in the Andean region. 
The Mexican cotton, G. hirsutum, appears to have 
arisen, as a species, in the south Mexican-Guatemalan 
region. Three varieties of this species were recognized at 
the time of the comprehensive work cited above, but 
Hutchinson (104) has since been able, with more ade- 
quate material, to recognize seven geographic races, five 
of which are cultivated in Mexico, asixth occurring only 
in the naturalized state in coastal Yucatan. The seventh 
race, ‘‘Marie-Galante,’* the most primitive of the group, 
occurs extensively in Central America, the West Indies 
and northern South America. 
The evidence that the distinct species, G. barbadense 
L., of South America, and G. hirsutum, have differen- 
tiated from a common ancestor while under cultivation 
is of very great interest. This implies a considerable an- 
tiquity for agriculture in both hemispheres and shows 
that definite conclusions concerning the relationships and 
origins of cultivated plants can be reached only after the 
most careful study. (104*, 105*, 188, 164, 191, 194, 203, 
206, 207, 210) 
[ 133 ] 
