CORN BREEDING 
31 
to plant a crossing block the next season. Pollen-parent plants were 
from the ear indicated by the ear-to-row test as most productive, and 
pistillate-parent plants were from the second-best ears. In all, each 
of 20 crosses were compared for periods of from one to three years. 
Of these, 18 were more productive and 2 were less productive than 
field-selected seed. The averages of the three to four crosses tried 
each year were more productive in every case, the average increase 
during the eight seasons covered being 5 bushels per acre. This in- 
crease was from the immediate progeny of the productive ears. How 
much of it would persist in later generations is not clear. 
Table 4. 
■Comparative yields of strains of com selected by the ear-to-row 
method in Ohio 
[Adapted from Williams and Welton (78, p. 
99)] 
Strain 
No. 
Years grown 
Average acre yields for years 
grown (bushels) 
Group 
Selected 
strain 
Check 
Increase 
above 
(+)or 
decrease 
below (—) 
check 
f 41 
42 
I 43 
f 95 
\ 96 
[ 97 
f 202 
203 
| 204 
I 205 
f 315 
\ 316 
I 317 
f 651 
\ 652 
I 653 
f 803 
J 804 
) 805 
I 806 
1907 and 1908 
77.45 
84.41 
83.68 
80.67 
82.67 
81.74 
64.72 
66.24 
62.51 
69.57 
88.62 
82.21 
83.03 
f 84. 16 
\ 82. 60 
[ 78. 91 
f 79. 84 
I 77. 37 
| 84. 60 
I 81. 22 
79.43 
79.43 
79.43 
74.43 
74.43 
74.43 
61.04 
61.04 
61.04 
61.04 
80.44 
80.44 
80.44 
79.03 
79.03 
79.03 
73.60 
73. 60 
73.60 
73.60 
-1.98 
No. 1 
_ .do 
+4.98 
No. 2 
do 
1909, 1910, and 1911 
do 
+4.25 
+6.24 
+8.24 
do 
1910 and 1911 . 
+7.31 
+3.68 
No. 3 
1911 and 1912 
+5.20 
+1.47 
+8.53 
+8.18 
No. 4 
do .. 
+1.77 
No. 5___ 
do 
il913 
+2.59 
+5.13 
+3.57 
No. 6 
ll914. 
-0.12 
+6.24 
+3.77 
+11.00 
+7.62 
The results of selecting for high and for low yield by the ear-to- 
row method at the Illinois Agricultural Experiment Station are of 
special interest (72). The experiment was begun in 1911 with a 
foundation stock of 990 ears. Seed of these ears was planted an ear 
to a row, and a composite lot of seed representing all 990 ears was 
planted in an isolated plat. This composite lot of seed constituted 
the foundation stock for a nonpedigreed strain that was propagated 
by simple mass selection. 
High-yield and low-yield selection plats of 40 ear rows each were 
planted in 1912 and subsequently. The remnant seed of the 40 
highest producing and the 40 lowest producing ears in the 1911 test 
was planted in 1912. After that, seed for the high-yield plat was 
selected from the 10 highest yielding rows of the preceding year's 
high-yield plat, whereas seed for the low-yield plat was selected 
from the 10 lowest yielding rows of the preceding year's low-yield 
