240 
Journal of Agricultural Research 
Vol. XXI, No. 4 
Of the 1,400 flowers which were examined in these four progenies, 459, 
or 33 per cent, showed no trace of a petal spot, and 165, or 12 percent, 
showed only the faintest discernible trace. Only 2 7 flowers were graded 
No. 3, and only 4 flowers were graded No. 4, the highest grade which was 
represented. None of the plants showed total absence of the spot in 
all 10 of the flowers which were examined, but on 1 plant each in pro¬ 
genies 1-3-2 and 3-2-4, 9 of the flowers showed no trace of the spot 
and the tenth flower showed only a faint trace (grade 1). 
The range of variation in petal spot development shown by the great 
majority of the 4-lock plants in 1920 (from o to grade 2) as compared with 
the normal development in Pima, the total absence of the spot in the 
Holdon variety of Upland cotton, and a somewhat subnormal develop¬ 
ment in the first generation of a hybrid between Pima and Holdon, are 
illustrated in Plate 54. 
The strongly hereditary nature of this variation is shown by the 
absence or slight development of the spot on all plants of the three 
progenies in 1919 and of the four progenies in 1920, whereas in both 
years all other stocks of the Pima variety which were growing in the same 
field and on which observation of this character was made showed a 
normal development of the petal spot. 
In regard to other color characters of the flowers, the petals on every 
plant in the 4-lock progenies in 1919 and 1920 had the full lemon-yellow 
color and the anthers had the full orange-yellow color which is charac¬ 
teristic of Pima, and in all other characters, excepting the greater number 
of 4-lock bolls and the absence or very slight development of the petal 
spot, no appreciable departure from the type of the variety was observed. 
The tendencies to increased lock number and to disappearance of the 
petal spot in these lines suggest that there may have been non-Egyptian 
“blood” in their remote ancestry. In most varieties of American Up¬ 
land cotton, as well as in the Hindi or “Weed cotton” of Egypt, 1 
4-lock bolls predominate and the petal spot is absent. There had been 
no opportunity for hybridization with either of these types subsequent 
to the origin of the Pima variety, but the possibility is not excluded 
that the factors in question were introduced through the Yuma ances¬ 
tors of the Pima parent, the Yuma variety having perhaps received 
them from its Mit Afifi progenitors. 
This association of very weak development or total absence of the 
petal spot with a relatively high percentage of 4-lock bolls suggests the 
existence of a negative correlation between petal spot and mean lock 
number. Within the 4-lock populations, however, no significant corre¬ 
lation could be detected; and in the second generation of a hybrid be¬ 
tween Egyptian and Upland cottons, of which 180 plants were grown at 
Sacaton in 1919, this pair of characters proved to be uncorrelated, the 
1 Cook, O. F. Hindi cotton in Egypt. U. S. Dept. Agr. Bur. Plant Indus. Bui. aio, 58 p., 6 pi. 1911. 
