ion of extreme wetness, it will be
divided into large masses, or clods, separated by wide cracks. A
subsequent wetting of the clods, which is not sufficient to expand it to
its former condition, will not entirely obliterate the cracks, and the
next drying will be followed by new fissures within the clods themselves;
and a frequent repetition of this process will make the network of
fissures finer and finer, until the whole mass of the soil is divided to a
pulverulent condition. This is the process which follows the complete
draining of such lands as contain large proportions of clay or of peat. It
is retarded, in proportion to the amount of the free water in the soil
which is evaporated from the surface, and in proportion to the trampling
of the ground, when very wet. It is greatly facilitated by frost, and
especially by deep frost.

The fissures which are formed by this process are, in time, occupied by
the roots of plants, which remain and decay, when the crop has been
removed, and which prevent the soil from ever again closing on itself so
completely as before their penetration; and each season's crop adds new
roots to make the separation more complete and more universal; but it is
only after the water of saturation, which occupies the lower soil for so
large a part of the year, has been removed by draining, that roots can
penetrate to any considerable depth, and, in fact, the cracking of
undrained soils, in drying, never extends beyond the separation into large
masses, because each heavy rain, by saturating the soil and expanding it
to its full capacity, entirely obliterates the cracks and forms a solid
mass, in which the operation has to be commenced anew with the next
drying.

Mr. Gisborne, in his capital essay on "Agricultural Drainage," which
appeared in the _Quarterly Review_, No. CLXXI, says: "We really thought
that no one was so ignorant as not to be aware that clay lands always
shrink and crack with drought, and the stiffer the clay the greater the
shrinking, as brickmakers well know. In the great drought, 36 years ago,
we saw in a very retentive soil in the Vale of Belvoir, cracks which it
was not very pleasant to ride among. This very summer, on land which, with
reference to this very subject, the owner stated to be impervious, we put
a walking stick three feet into a sun-crack, without finding a bottom, and
the whole surface was what Mr. Parkes, not inappropriately, calls a
network of cracks. When heavy