100 
and potassic in a third? "We have tried potassium chlorid and sodium chlorid side 
by side. Potassium increases the yield threefold and sodium not at all. To be sure 
we have been studying Illinois soils for only two years, and we have made only a 
mere beginning in that great State, but we are making use of soil chemistry, soil 
physics, soil bacteriology, pot cultures, field experiments, and in fact every method 
or agency which promises to aid us, and we hope to rapidly obtain much more com- 
plete knowledge of Illinois soils than we now have. 
Bulletin No. 68, just received from the Florida Experiment Station, contains forty 
chemical analyses of the ordinary very sandy loams upon which nearly all of the 
pineapples produced in that State are grown. In commenting upon these soils, the 
authors say:" "Few of the soils would be able to produce more than two or three 
crops of pineapples, if all the plant food present were available." Probably these 
sandy loams should be considered as abnormal soils, but there are actually all grada- 
tions between these sandy soils and the heaviest clays or the most peaty swamps. 
Where shall we draw the line between the soil whose fertility can be reduced so as 
to affect the crop yield and the soil whose supply of fertility "will be indefinitely 
maintained." 
The conclusions of the Bureau of Soils reported in Bulletin No. 22 were based in 
part upon the fact that no special correlation was found between ordinary orop 
yields and the chemical composition of an aqueous extract of the soil and in part 
upon a cursory examination of the literature bearing upon the subject. I say a cursory 
examination because of the large amount of existing data which appear to have been 
overlooked. 
For example, in the bulletin from the Bureau of Soils it is suggested that the appli- 
cation of plant food is usually of little or no value, provided a proper rotation is 
practiced, and the results obtained from wheat grown continuously and from the 
four-year rotations of wheat, roots, barley, and fallow, which have been carried on 
at Rothamsted during fifty years are cited as proof. The statement l > is made that 
"the yield of wheat grown continuously without manure for fifty years has been 
reduced from 33J bushels, the average maintained on the best fertilized plat, to 15 
bushels." 
One would at first suppose this statement were a misprint, We might almost in 
truth make the opposite statement, namely, that by the use of farm manure the yield 
of wheat grown continuously has been increased from 13£ bushels, the average main- 
tained on the unfertilized plat, to 33^ bushels. It is not the reduced yield from crop- 
ping without manuring that is noteworthy, but it is increased yield due to the appli- 
cation of plant food. The unmanured plat never produced 33^ bushels. 
Wheatgrown continuously at Rothamsted c {bushels per acre) . 
Harvest year. 
Without 
manure. 
With farm 
manure. 
Differ- 
ence. 
1843 
dl8 
H18 

1844 
15 
23 
18 
17 
15 
19 
16 
16 
21 
32 
27 
30 
26 
31 
28 
30 
6 
1845 
9 
1846 . . . 
9 
1847 
13 
1848 
11 
1849 
12 
1850 
12 
1851. . . 
14 
17 
28 
11 
1884 
13 
15 
9 
15 
10 
12 
14 
14 
32 
40 
36 
36 
38 
10 
13 
48 
19 
L885 
25 
1886 
27 
1887 
20 
1888 
28 
1889 . 
28 
L890 . 
29 
1891 . . 
34 
13 
39 
26 
13| 
33* 
20 
a Florida Station Bui. 68, p. 691. 
i>v. s. Dept. Agr., Bureau of Soils Bui. 22, p. 65. 
<■ Agricultural investigations at Rothamsted, V . 8. Dept Agr., Office of Experiment Stations Bui. 22, 
PI. I, between pp. 1 16 and 117. 
''Supposed yields for 1843. 
