Experimental Crops at Rothamsted. 
109 
the drains were running, and it was supposed that the ground 
was quite saturated. It was, indeed, so wet that it was neces- 
sary to lay down boards for the men to stand upon whilst 
working. 
Table v., overleaf, shows the percentages of moisture in the 
different samples of soil ; bringing together — first, the results 
for the three plots during the drought ; second, those for the 
three plots when the land was saturated ; and lastly, the same 
results arranged for the convenient comparison of the percentages 
in the dry state and the wet state, and showing the difference 
between the two, for each plot separately. 
It will be obvious that the amount of water at the different 
depths in July, 1868, after about three months of great deficiency 
of rain, and the growth of a crop then approaching ripeness, 
must, in the main, be dependent on the supplies accumulated 
during the previous winter and early spring. But it is affected, 
to a greater or less depth from the surface : by any difference of 
texture and power of absorption, the result of previous culti- 
vation, manuring, and cropping ; by the influence of the pipe- 
drains, which are at a depth of about 30 inches ; also, by the shade 
of the crop on the one hand, lessening evaporation from the soil 
itself, and on the other, by the requirements of the growing crop 
increasing, according to its amount, the exhalation through the 
plants themselves, and the consequent pumping out of the stores 
within the soil. 
The soil of Plot 3, which had received no manure and pro- 
duced little root (tending to disintegrate the soil and increase its 
absorptive surface), which had comparatively little shade from the 
growing plants, preventing surface evaporation, and whose crop 
would exhale comparatively little, is seen to retain a somewhat 
less percentage of water than either of the others within 3 inches 
of the surface, but more than either within the next 9 inches. 
In it, as in the others, the percentage of moisture increased gra- 
dually from that point downwards, until obviously affected by the 
action of the pipe drains. 
The soil of Plot 2, which had then been manured with 
14 tons of dung per acre per annum for twenty-five years in suc- 
cession, notwithstanding the greater requirements of the crop, 
retained rather more moisture than the unmanured soil within 
3 inches of the surface ; a result partly due, perhaps, but 
not wholly, to more shade. But, from that point downwards, 
doubtless influenced by the requirements of the crop, the dunged 
soil retained less at every stage (excepting the lowest) than the 
unmanured. 
The soil of Plot 8, manured annually with mineral manure 
and ammonia-salts, and yielding pretty uniformly a heavier crop 
