ediate
channel we may find gravels, sand, and loam (fig. 255, 2 2')
deposited by the river flowing in its present bed. These are
_recent_ fluviatile or alluvial deposits. At some distance from
the present bed of the river, and at a higher level, we may find
other sands and gravels, quite like the recent ones in character
and origin, but formed at a time when the stream flowed at a higher
level, and before it had excavated its valley to its present
depth. These (fig. 255, 3 3') are the so-called "_low-level_
gravels" of a river. At a still higher level, and still farther
removed from the present bed of the river, we may find another
terrace, composed of just the same materials as the lower one,
but formed at a still earlier period, when the excavation of
the valley had proceeded to a much less extent. These (fig. 255,
4 4') are the so-called "_high-level_ gravels" of a river, and
there may be one or more terraces of these.

[Illustration: Fig. 255.--Recent and Post-Pliocene Alluvial Deposits.
1, Peat of the recent period; 2, Gravel of the modern river:
2', Loam of the modern river; 3. Lower-level valley-gravel with
bones of extinct Mammals (Post-Pliocene); 3', Loam of the same
age as 3; 4. Higher-level valley-gravel (Post-Pliocene); 4',
Loam of the same age as 4; 5. Upland gravels of various kinds
(often glacial drift); 6, Older rock. (After Sir Charles Lyell.)]

The important fact to remember about these fluviatile deposits
is this--that here the ordinary geological rule is reversed. The
high-level gravels are, of course, the highest, so far as their
actual elevation above the sea is concerned; but geologically the
lowest, since they are obviously much older than the low-level
gravels, as these are than the recent gravels. How much older
the high-level gravels may be than the low-level ones, it is
impossible to say. They occur at heights varying from 10 to 100
feet above the present river-channels, and they are therefore
older than the recent gravels by the time required by the river
to dig out its own bed to this depth. How long this period may
be, our data do not enable us to determine accurately; but if
we are to calculate from the observed rate of erosion of the
actually existing rivers, the period between the different
valley-gravels must be a very long one.

The lowest or recent fluviatile deposits which occur beside the
bed of the present river, are referable to the Recent period,
as they contain the remain