6 2 
Journal of Agricultural Research voi.xxvii.no. r 
The results shown in Table VI indicate that Sunrise kafir and Red 
kafir differ by a single main factor for seed color. The factors influencing 
color in the Sunrise X Red kafir cross may be designated as rr for Sun¬ 
rise kafir and RR for Red kafir. The seed color of the cross is due to 
the combination Rr in the F x and the segregation of the Rr factor pair 
in the F 2 generation in the monohybrid ratio of 3:1. While it seldom is 
advisable to formulate a factorial hypothesis on the behavior of only 
287 F 2 individuals, in this instance it seems safe to do so, as the data 
on other crosses, White kafir X Red kafir and White kafir X Sunrise 
kafir, reported in this paper, bear out this theory. Sunrise kafir and 
White kafir have a similar seed color and the color results obtained in 
White kafir crosses should be the same as those in which Sunrise kafir 
is used. 
WHITE KAFIR X RED KAFIR 
A cross was made in 1919 between White kafir and Red kafir, the 
White kafir being the pistillate parent. Two F 2 rows were produced 
in 1921 from seed of selfed F 4 heads. The segregation of seed-color in 
the plants in these two head rows is shown in Table VII. 
Table VII. —Distribution of seed color in the F 2 generation of the cross White 
kafirXRed kafir , grown in IQ2I 
Row number. 
Total 
number of 
plants. 
Plants 
with red 
seeds. 
Plants 
with white 
seeds. 
I. 
202 
198 
150 
151 
52 
47 
2. 
Total. 
400 
301 
300 
99 
100 
Expected (3:1 ratio). 
Deviation, i±5*84. , = 0.171. 
P. E. 
Six selfed heads from each of the two F 2 rows of this cross were used 
for sowing head rows in 1922. These 12 heads were selected for other 
characters than color of seed, the selection not being random, but seed 
color was recorded both for the parent heads and the F 3 progenies. 
The data obtained in the F 3 generation are shown in Table VIII. 
Six of the F 2 heads were classed as red and produced nothing but red- 
seeded progenies, with a total of 659 individuals. The three white-seeded 
F 2 heads produced a total of 364 white-seeded individuals. The three 
parent heads classified as light red obviously were the het rozygotes, pro¬ 
ducing red- and white-seeded plants in a ratio fairly close to 3:1. As 
noted, the parent heads were not random selections, so that the fact that 
the progenies did not occur in a 1:2:1 ratio of dominants, heterozygotes, 
and recessives is immaterial. Especially is this true in view of the fact 
that the heterozygotes can be determined by their intermediacy. 
