66 
INVESTIGATIONS ON KOTHAMSTED SOILS. 
there were no drainpipes to tap the soil higher up. It is also to be 
noted that the samples were taken in a fallow year (the plat being 
wheat cropped and fallowed in alternate years), and that, therefore, 
there had been no growing crop to assimilate the surface-formed 
nitrates. On the other hand, the sampling was carried out in July, 
when summer nitrification was very far from being completed, while 
the samples from the Broadbalk field now under consideration were 
taken in October, i. e. , nearly three months after the crop would have 
been actively engaged in taking up nitrogen, though nitrification 
would have been, during these three months, in active progress. 
On the whole it would seem, in the light of the 1893 samples, that 
the larger quantit} r of nitric nitrogen found in the lower depths of 
Hoos field was partly related to the absence of a recently growing 
crop, but more particularly to the absence of artificial drainage. 
In another series of samples, taken toward the close of 1882 in 
Agdell rotation field, after wheat following fallow, 1 the average quan- 
tity of nitric nitrogen in the nine depths below the first 27 inches was 
1.3 pounds per acre. Here again, though there had been a growing 
crop, there was no artificial drainage, and its absence coincides with 
the greater quantity of nitric nitrogen than we find in the lower 
depths of the Broadbalk plat unman ured with nitrogen. . The six 
consecutive depths below the drainpipes on plat 5 of the Broadbalk 
field in 1893 contained 6.05 pounds of nitric nitrogen per acre; the 
corresponding depths of Hoos field (wheat fallow, sampled in July, 
L893) contained 15.49 pounds per acre. 
From the quantity of nitrates actually found in the waters collected 
from the drainpipes of the Broadbalk field during some years it is 
estimated that sometimes nearly 17 pounds of nitrogen per acre have 
passed away in drainage from the wholly unmanured plats, and over 
18 pounds per acre from plat 5, which we are at the moment consider- 
ing. Bearing this in mind, and also the difference of conditions in 
the o1 her fields, there can be little doubt that the larger quantity of 
aitric nitrogen found in their lower depths, and the absence of any 
marked break between the third and fourth depths is due to down- 
ward percolal ion. 
Quite apart from other evidence the examination of the lower depths 
of the ammonia-manured plats, which will be considered presently, 
shows very clearly that the drainpipes are very far from being able 
to remove completely the nitrates produced by the oxidation of 
unused ammonium salts and of crop residues in the soils fed with such 
salts, for we find Largely increased nitric nitrogen, despite the drain- 
pipes, all the way down to 90inches. It seems, therefore, only reason- 
able to assume thai even on the unmanured plats, as nitrates occur to 
so large an extent in the actual drainage water collected from the 
pipes, a considerable quantity must be carried down in the undis- 
i Journal of the Chemical Society of London, Transactions, 47 (1885), p. 406. 
