130 
In this experiment five sections of 30 plats each, A,B, C, D,and E, have been laid 
out on land apparently uniform in formation and previous, treatment. Beginning 
with plat 1, every third plat up to X<>. 28 has been left continuously unfertilized. 
The experiment is defective in that there was not a final unfertilized plat — No. 31 — 
hut this was omitted because of limitation of suitable land. The present season gives 
the tenth annual crop on sections A and C, the ninth on D and E, and the eighth on 
B, of winch records have been kept, the rotation beginning in 1894. 
Diagram II shows the annual yield and increase in total produce — corn and stover, 
wheat, oats and straw, and hay — for the whole period on section C, and Diagram III 
shows the same for the average of the five sections. In the ten years covered by two 
complete rotations this would give two crops of each of the cereals and four of hay on 
each section, the rotation being corn, oats, and wheat, one year each, and timothy 
and clover mixed, two years. ( )f late years the clover has frequently failed to stand 
over winter, in which case it has been plowed under and soy beans have been grown 
instead, the beans being followed by millet or by another crop oi beans. This, how- 
ever, does not affect the present discussion. 
The broken line represents the level of the unfertilized yield, as found by carry- 
ing the line from one unfertilized plat to the next, according to the method I have 
explained, and it will be seen that there has been a marked variation in the yield of 
some of these plats. No. 28 is especially noticeable in this respect. The land upon 
which this section i:; located was all in one field, as originally purchased by the sta- 
tion, and no difference Mas observed in its character until the yield of the first crops 
3500 
3000 
2500 
■ |J 
J" 
. J- ■- j-jI 
I 
I 
c 
2000 
1500 
I 
B 
cr 
a 
PLAT NO. 
i 
2 
3 
4 
5 
6 
7 
8 
9 
10 
II 
1'. 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
26 
27 
28 
29 
30 
Diag. III.— Ten-year average yield and increase on the five sections, Ohio Experiment Station. 
grown in preparation for this test came to be examined, when the sudden rise in the 
yield of plat 28 caused inquiry to be made regarding the previous history of the field. 
This inquiry revealed the fact that a lane had once occupied the space now covered 
by plat 28 and parts of the adjacent plats 27 and 29. This lane had been abandoned 
some ten years or more before the land came into possession of the station and was 
thrown into the field, yet the yield of corn this year, 1903, twenty or more years after 
the lane had been abandoned, is nearly double the best yield on any other unfer- 
tilized plat in the series. A glance at the diagram will show that a computation of 
increase on the basis of the general average of all the unfertilized plats, as shown by 
the line AB, would have given enormously exaggerated yields to the plats at this end 
of the test, while, had we adopted the formerly common method of leaving but one 
or two unfertilized plats and No. 28 had happened to be one of them, and had our 
work been limited to this one section, we should have had results absolutely mislead- 
ing and worthless. The line AC would then have indicated the supposed average. 
It should be explained that plate 17, 21, 23, and 24 receive the same quantities of 
nitrogen, phosphorus, and potassium, the nitrogen being conveyed in different 
carriers. Plats 11, 26, 27, and 29 also receive the same quantities of each of the three 
elements, the phosphorus being carried in raw bone meal to plat 26, in dissolved 
bone black to plat 27, and in basic slag to plat 29. The first four plats now receive half 
the quantity of nitrogen and twice the quantity of phosphorus that is given to the 
other four, but previous to 1899 the eight plats received the same quantities of each of 
the three elements. 
