Aug. 26, 1918 
Variation and Correlation in Wheat 
379 
The weight classes for the individual seeds planted, number of plants 
in each class harvested, and the average yield, in decigrams, of kernels 
per plant are given in Table XII. This gives the same data with respect 
to yield from kernels of different weights as is given in the correlation 
tables but in more direct form. That increase in weight of seed planted 
was not consistently followed by increased yield of kernels is evident. 
Table XII .—Weight classes of seeds planted and average yield of kernels per plant 
Weight 
classes of 
individual 
seeds 
planted. 
1914 
1915 
1916 
1917 
Number 
of plants 
harvest¬ 
ed. 
Average 
yield per 
plant. 
Number 
of plants 
harvest¬ 
ed. 
Average 
yield per 
plant. 
Number 
of plants 
harvest¬ 
ed. 
Average 
yield per 
plant. 
Number 
of plants 
harvest¬ 
ed. 
Average 
yield per 
plant. 
Mgm. 
Dgm. 
Dgm. 
Dgm. 
Dgm. 
6. 
25 
. 16.38 
8 . 
4 
18. 28 
15 
11. 25 
41 
21. 77 
29 
3 
$• 67 
48 
12.84 
34 
26.21 
1 S::::::: 
4 
7.00 
7 
31.07 
59 
14.89 
39 
28.01 
18. 
20 
4. 20 
8 
32.50 
36 
15. 88 
33 
28. 23 
20. 
23 
6. 74 
35 
37. 21 
25 
16. 23 
39 
26. 50 
22. 
28 
7. 64 
52 
42.98 
38 
15-91 
37 
27-93 
24 . 
27 
8.26 
64 
38- 98 
68 
18. 95 
36 
26.08 
26. 
15 
7 * 93 
45 
46. 28 
77 
17. 89 
38 
26. 84 
28. 
7 
6-43 
11 
33-36 
62 
18. 52 
36 
29.08 
3 ©. 
8 
9. 00 
20 
38. 00 
44 
16.47 
32 
31- 22 
32....... 
6 
10.33 
47 
38. 88 
17 
17-25 
25 
32.70 
8* 82 
38 
35 - <0 
11.14 
21 
39. 88 
45 
*8 . 
18 
10 
42. 50 
47 
20. 86 
10. 81 
46 
39. 02 
30 
8. 20 
65 
38. 42 
16 
26 
6. 62 
49 
45.97 
6 
22. 50 
8. 50 
38. 79 
A& . 
5 
6. 60 
61. 50 
S- SO 
8 
57 - 5 ° 
| 
Regression for yield of kernels' in each of the four years is shown in 
figures 5, 6, 7, and 8. Regression for yield was consistently greater 
in 1916 (fig. 7) for the seeds up to 24 mgm. than for the larger seeds. 
Apparently increase in amount of endosperm in 1916 and 1917 up to 
the weights of seeds indicated gave more uniformly proportionate in¬ 
creases in yield than were given by increases in endosperm beyond these 
amounts. 
Very similar to the correlations between weight of seeds and yield 
of kernels in each of the four years are those between weight of seed 
and total length of culms, total length of spikes, number of culms, total 
yield, yield of straw and number of kernels per plant. 
The correlation between weight of seed and plant characters at matu¬ 
rity in 1916 and 1917 was, for height of tallest culm, low and variable; 
for average height of culms and average length of spikes, low but less 
variable; and for total length of culms, total length of spikes, number 
of culms, total weight, and yield of grain and straw, medium, with little 
fluctuation. The only significant difference in 1916 and 1917 between 
coefficients for weight of seed planted ahd characters at maturity is 
