88 
INVESTIGATIONS ON ROTHAMSTED SOILS. 
per annum, or only about the same quantity as was found in the 
drainage of the plats receiving no nitrogenous manure. 
There was found in the drainage water from the sodium-nitrate 
plat a much larger quantity of soda than in the drainage from the 
other plats, the excess being more than sufficient to combine with the 
nitric acid present. It is often, and apparently without justification, 
stated that nitrate of soda robs the soil of lime, on the ground that 
the unused excess which passes away in the drainage water passes 
away as calcium nitrate. But, on the Rothamsted soils at all events, 
which are naturally abundantly supplied with lime, such does not 
appear to be the case. It is, on the contrary, the ammonium salts 
which behave nefariously as regards the lime of the soil, and there- 
fore it appears inadvisable in practical agriculture to use ammonium 
salts in place of nitrate of soda, except on soils which contain a fair 
quantity of lime. If ammonium salts are used persistently on soils 
poor in lime, care must be taken to restore lime from time to time* to 
guard against this form of loss. 
The impoverishment of the surface soil in lime under the continued 
application of large quantities of ammonium salts appears to be indi- 
cated by the following analyses which, at the suggestion of Sir John 
Lawes, I recently made of some of the 1893 samples of the Broadbalk 
soils in my possession : 
Table 46. — Broadbalk wheat soils, samples collected in October, 1893 — Lime in 
soils and subsoils. 
Depth. 
Plat 3, continu- 
ously u n m a - 
nured. 
Plat 5, full min- 
eral manures; 
no ammonium 
salts. 
Plat 7, full mineral 
manures, with 
400 pounds am- 
monium salts 
per acre per an- 
num. 
Per cent. 
Pounds 
per acre. 
Per cent. 
Pounds 
per acre. 
Per cent. 
Pounds 
per acre. 
First (1 to 9 inches) 
2.464 
.481 
.638 
(53,881 
12,849 
17,820 
2.766 
.683 
.582 
71,711 
18,245 
16,246 
2. mo 
.739 
.694 
53, 407 
19,741 
19,373 
Second CIO to 18 inches) 
Third (19 to 27 inches) 
Both plat 5 and plat 7 receive a considerable yearly supply of sul- 
phate of lime in the form of superphosphate. Plat 5, as we should 
expect, is much richer in Lime in the first and second depths than the 
nnmanured plat; hut plat 7, which receives annually ammonium sal ta 
as well as superphosphate, is in the surface soil far poorer. 
The difference in lime between the surface soils of plats 5 and 7 is 
no less than 1.8,304 pounds per acre. The Lime supplied to both plats 
in manure has been the same in quantity; and if the soils originally 
contained Like proportions of Lime, the loss of lime to the surface soil 
Owing to the ammonium salts has been about 360 pounds per acre per 
ami um. 
