428 Journal of Agricultural Research voi.xxi.No 7 
tests for two seasons. The length of kernel was one of the characters 
studied. The average length of the kernels in the high-yielding and 
low-yielding lots were 0.42 and 0.41 inches, respectively, and the yields 
were 72.5 and 54.8 bushels per acre, respectively. Concerning this point 
the authors conclude that length of the kernels was a negligible factor 
in the yields. 
Olson, Bull, and Hayes 1 found that smooth and medium smooth ears 
of Minnesota No. 13 com slightly outyielded rough ears in ear-to-row 
tests conducted at the Minnesota Agricultural Experiment vStation. 
PLAN OF THE EXPERIMENT 
An experimental project was put under way at the Kansas Agricul¬ 
tural Experiment Station in 1916 to secure some data concerning the 
relation of length of kernel to the yield of corn. Commercial White, a 
medium large variety well adapted to growing on productive soils in 
eastern Kansas, was used. Three 40-ear lots of corn were selected. 
One lot was made up of ears with kernels that were relatively short, one 
of ears with kernels of maximum length, and the third of ears with kernels 
of medium length. The latter group was practically intermediate 
between the short- and long-kerneled ones. Since the degree of indenta¬ 
tion is usually correlated with length of kernel, this character was taken 
as an index by which to classify the various ears. Only ears that were 
almost smooth or “dimple-dented” were selected for the short-kerneled 
group, while only ears that were sufficiently indented to have kernels 
with a chaffy crowm or with at least an indication of chaffiness at the 
crown were used for the rough- or deep-kerneled group. In this latter 
group the endosperm as a rule was not completely filled. The kernels 
of the intermediate group were “ wrinkle-dented,” that is, the indentation 
was sufficient to cause a wrinkling of the epidermis of the kernel over 
the indented portion. The ears shown in Plate 81 are typical for the 
group, while Plates 80 and 82 show typical ears of the long- and sliort- 
kerneled groups, respectively. These groups are designated as “rough” 
and “smooth,” respectively, while the group with kernels intermediate 
in length is known as “ medium.” This last group was practically identi¬ 
cal in type and length of kernels with the mean for the Commercial White 
variety. Kernels of the three types are shown in Plate 83. 
The seed for the various groups or types of ears was continuously 
selected. Smooth ears were selected each season from the progeny of 
smooth ears, rough ears from the progeny of rough ears, and medium 
ears from the progeny of medium ears. 
In selecting the seed ears, care was taken to secure, if possible, ears 
that would conform to what is ordinarily considered the standard in 
seed com. The ears were well-developed, with uniform kernels and in 
good seed condition. 
1 Olson, P. J., Bull, C. P., and Hayes, H. K. ear type selection and yield in corn. Minn. Agr. 
Exp. Sta. Bui. 174, p. 28. 1918. 
