Experimental Crops at Rothamsted. 
115 
The unmanured soil and subsoil, comparatively little disturbed 
and disintegrated by the permeation and the decomposition of 
roots, and not at all by the action of manures, would offer less 
surface and absorb less water, and they are seen to retain less 
than those of either of the manured plots. The soil and sub- 
soil of the artificially manured plot would be affected by the 
permeation not only of more roots, but of the solution of the 
manures or of some of their products of decomposition, — by 
the latter especially in the lower layers. But it is the dunged 
plot, with its vast accumulation of organic matter near the sur- 
face, and its finely divided and dissolved products of decomposi- 
tion permeating to a greater or less depth beyond, and, doubtless, 
a considerable development of root, that is seen to possess 
the greatest power of retention of moisture, especially near the 
surface. 
Taking the figures relating to the saturated soils as they stand, 
the artificially manured plot retained 153 tons, and the dunged 
plot 214 tons more water per acre, to the depth examined, than 
the unmanured — amounts which represent, respectively, about 
1^, and more than 2 inches of rain. Or, if we take the difference 
between the amounts retained in the dry and the wet condi- 
tions, the dunged soil shows a still greater excess of absorption 
when saturated, both compared with the unmanured, and with 
the artificially manured soils. Further, the details show that the 
dunged soil, when saturated, retained, within 12 inches of the 
surface, an excess of water which would be equivalent to about 
1^ inch of rain more than that held to the same depth on either 
of the other plots. 
In connection with this interesting fact, it may be mentioned, 
that whilst the pipe-drains from every one of the other plots in 
the experimental wheat-field run freehj, perhaps on the average 
four or five times annually, the drain from the dunged plot seldom 
runs at all more than once a year : indeed, it has not with cer- 
tainty been known to rnn, though closely watched, since about 
this time last year. At first it was thought that there must be 
some stoppage, or some fault in the levels. Accordingly, the 
soil was opened in various places, but was found to be far from 
saturated down to the range of the drains. It was then con- 
cluded that the result was due to the greater power of absorption 
and retention of moisture by the dunged soil near the surface ; 
and even supposing the figures above given should exaggerate 
the difference actually occurring, there would still be a wide 
margin remaining, sufficient to account for the fact of no water 
reaching the drains excepting under the influence of an unusually 
large and continued rainfall. Such a fact as the one here 
recorded is obviously of great interest and significance. Whether 
I 2 
