July is, 1914 
Mutation in Egyptian Cotton 
295 
In 1909 a progeny row of 16 plants was grown at Sacaton from seed 
of this selection and showed a remarkable degree of uniformity in the 
plants and the fiber. In 1910 a half-acre plat was planted with the seed 
from this progeny row. During subsequent years the plantings have been 
gradually increased until, in 1913, 200 acres were grown on the Pima and 
the Maricopa Indian Reservations in southern Arizona (PI. XVII). 
CHARACTERS OR THE GITA COTTON 
Main stem rather slender, with relatively short internodes, its first 
fruiting branch usually borne at the ninth or tenth node; vegetative 
branches slender, developing late, usually remaining shorter than the 
main stem; fruiting branches with rather short internodes and numerous 
bolls; leaves rather small, those of the main stem usually 3-lobed, the 
lobes deep; involucral bracts triangular ovate, separate or nearly so to 
the base; bolls short, plump, abruptly narrowed to the blunt apex; seeds 
large, having usually one-third to one-half of the surface covered with 
fuzz; fiber about i^g- inches long, somewhat darker colored than in the 
Yuma and Pima varieties. 
THE MUTABILITY OF EGYPTIAN COTTON 
Four distinct varieties of Egyptian cotton—Yuma, Somerton, Pima, 
and Gila—have arisen in the course of plant-breeding work in Arizona. 
Three of these were derived directly from the Mit Afifi, while the Pima 
variety is an offshoot from the Yuma. Each of these varieties originated 
with a plant which was very different from the parent stock, and the dis¬ 
tinctive characters have continued to be expressed with a high degree of 
uniformity during several generations. These facts seem to warrant the 
conclusion that the varieties mentioned are derived from mutants com¬ 
parable with those of Oenothera Lamarckiana as described by De Vries. 1 
The abrupt and distinct change of expression of characters in the parent 
individuals places the phenomenon outside the range of mere fluctuation, 
while the uniformity with which the new 2 characters have been ex¬ 
pressed in each subsequent generation makes it wholly unlikely that 
these forms are an immediate product of hybridization. 
As to the varieties which have originated in Egypt, while their history 
is much less completely known than that of the Arizona varieties, the 
data at hand point strongly to the conclusion that they also have been 
derived from mutants. 
1 The presence each year in the field plantings of the new varieties of Egyptian cotton of a small percentage 
0 f"off-type” plants is readily explained by the fact that the individuals which gave rise to the varieties 
were not protected from cross-pollination by surrounding plants of different character. De Vries (1909, 
v. 1, p. 275) states that before he resorted to bagging and self-pollinating the flowers of his mutants "these 
strains exhibited a very high degree of, though not an absolute, constancy.” 
2 It is highly improbable that any of the characters exhibited by these mutants are new in the sense 
of having been absent in all lines of the ancestry. The history of Egyptian cotton indicates that more than 
one species of Gossypium has contributed to the formation of the type. It is therefore probable that a large 
share of the characters which are possessed by the different members of this genus have been transmitted in 
the Egyptian complex and may come to expression in its mutants. 
4753i°—14-4 
