78 BULLETIN 754, U. S. DEPARTMENT OF AGRICULTURE. 
from the same cross may be readily compared. Ear No. 1564 had a 
correlation of 0.260+0.06. A correlation of this magnitude would 
be expected to occur as the result of chance once in 267 times, but 
since it occurs in a group of four, it would be expected as the result 
of chance once in 67 times. ’ 
Ear No. 1704 had a correlation of 0.3500.078, which is 4.5 times 
the probable error, and should occur as the result of chance once in 
415 times, but since it is but 1 in a group of 10 individuals, the 
odds against its chance occurrence are 42 to 1. We may with some 
propriety conclude that the correlations found on these two ears are 
accidental. The results, then, as shown by the third-generation ears 
support the assumption that this hybrid is homozygous for the color 
factor C: ; 
CORRELATION IN THE THIRD GENERATION OF THE HYBRID 
DH 237. 
Two ears were selected and grown from the hybrid Dh 287. 
These two ears are shown in Table XXXII, pedigree Nos. 1129 and 
1130. Ear No. 1129 was the result of self-pollinating a plant of the 
hybrid Dh 287, and ear No. 1130 is a second ear from this same 
plant. The male parent of this latter ear, when self-pollinated, 
produced ear No. 1135. Ear No. 1129 had the exact degree of corre- 
lation which is expected on a 8-1-1-3 coupling, 0.766 0.026, while the 
second ear of the same plant, No. 1130, had a correlation of 0.794 
+0.025. The male parent of ear No. 1130 had a correlation of 0.774 
+0.029. These two ears are very close approximations of the 0.766 
correlation which is expected on a 3-1-1-38 coupling. There were, 
of course, four classes of seeds on each of these two ears. The dif- 
ferent classes were planted separately. 
Crosses were made between plants from the white waxy seeds and 
plants from the colored horny seeds from the same ear. These 
crosses were made, using white waxy plants as the female parent, 
and also as the male parent. Self-pollinated ears were also secured 
from all the plants to make sure that they were of the type expected. 
The self-pollinated colored horny ears, however, are the only self- 
pollinated ears that would have all four classes of seeds. The three 
ears obtained from self-pollinated plants grown from colored horny 
seeds of ear No. 1129 are shown in Table XX XVIII. 
The correlation for these three ears is 0.770+0.017. This cor- 
relation is certainly a close approximation of the expected 0.766, but 
upon examining the three ears that make up the total we find the 
difference between the extremes to be 0.3840.06, a difference of 
more than six times the probable error, which can hardly be ascribed 
to chance fluctuation. The ear with the lowest correlation (No. 
1796, correlation 0.450+-0.063) deviates from the expected 0.766 by 
