July l, 1926 Productiveness and Deleterious Characters in Com 
87 
Inasmuch as these data are based on comparisons between ears all of winch 
were suitable for seed, the preponderance of evidence in certain cases seems 
convincing in spite of the fact that the determining differences in yield are small. 
There is every indication that selection on the basis of production (weight of 
ear in this case) is of value. Likewise it is indicated that it is preferable to 
obtain production by adding to the length rather than to the circumference of 
the ears, and that smoother, fewer-rowed ears with a lower shelling percentage 
than the standard show type are inclined to be the better yielders. 
Kiesselbach 4 stated in substance that in a six-year comparison 
the long, slender, smooth ears surpassed all other kinds, including 
the long, large, rough ears; the short, large, rough ears; the short, 
slender, smooth ears; and unclassified seed of Nebraska White Prize, 
the variety from which the selections were made. He does not state 
but shows in an illustration that the most productive type had mark¬ 
edly fewer rows than the ordinary Nebraska White Prize. He also 
shows in the same bulletin that hybrids producing smooth, few-rowed 
ears yielded more than hybrids producing ears with a larger number 
of rows and rougher kernels. 
Several experiments were conducted at different points in South 
Carolina and Florida. The performances of groups of ears differing 
in the number of rows of kernels were compared at Lykesland, S. C., 
in 1914, 1916, and 1917; at Brooksville, Fla., in 1917; and at Darling¬ 
ton, S. C., in 1915 and 1916. In the experiments at Lykesland the 
ears with different numbers of rows were classified further on the 
basis of angularity of kernel in 1914 and on the basis of kernel size 
in 1916. Additional studies of the relation of angularity of kernel 
to yield were made within a variety at Lykesland in 1914 and 1915, 
and at Darlington in 1915 ? and within two series of varietal crosses 
at Darlington in 1914. Summaries of the data from these experi¬ 
ments are given in Tables III, IV, V, VI, and VII. 
NUMBER OF KERNEL ROWS AND YIELD 
Data on the relation of number of kernel rows to yield in the 
Roger and Williamson varieties are shown in Table III. These are 
typical nonprolific varieties, the ears being rough, with compara¬ 
tively many kernel rows and tightly spaced angular kernels. Three 
groups of ears in each variety were compared in 1915, and six groups 
in the Roger variety were compared in 1916. 
Table III.— Data on the relation of the number of kernel rows to yield in the 
Roger and Williamson varieties at Darlington , S. C., in 1915 and 1916 
\ 
Points cons dered 
1915 
1916 
Williamson 
Roger 
Roger 
| 12 
16 
1 
1 14 
] 
Number of rows on seed ears_ 
{and 
and 
\ 20 
12 
<and 
f 18 
10 
12 
14 
16 
18 
* 20 
1 14 
18 
J 
1 16 
J 
Number of perfect hills grown. 
602 
506 
530 
462 
414 
453 
730 
751 
763 
778 
727 
700 
Relative yields.... 
108 
107 
97 
106 
94 
88 
98 
100 
105 
96 
94 
95 
The yields were inversely proportional to the number of kernel 
rows in both varieties in 1915. The highest yield in 1916 was from 
« Kiesselbach, T. A. corn investigations Nebr. Agr. Exp. Sta. Research Bui. 20, 151 p. 1922. 
