PLEISTOCENE CLASSIFICATIONS.
269
suggestion (inverting the system of Penck) that the glacial
episodes are not to be correlated with the river terraces and
the marine base-levels, but with the gaps between the terraces.
The Deperet system is essentially based upon the supposition
of uniform changes of sea-level, but where we are presented
with differential earth movements, the problem of applying it
to local conditions becomes very complicated.
As is well known, the difference between the emergence of
Northern Ireland and Scotland, and the submergence of England,
has been a very considerable one, even since Neolithic times.
There is also evidence of differential movement between
the Chellian and the Mousterian epochs. In the Somme Valley
the excavation of the channel had been nearly completed prior
to late Chellian times. At about that date submergence set in,
and this was apparently accompanied by silting-up of the valley,
as Acheulian river deposits are found about 80 feet above the
bed of the Chellian channel. The history of the Thames valley
is quite different. Here it is the older basement gravels of the
Taplow (=50-foot = Middle) Terrace which are of Late Chellian
age. Emergence and rejuvenation then set in (in place of sub-
mergence) and a deep channel was cut which passed away into
what is now the North Sea (2). The buried channel of the
Thames comes in still later somewhere near the boundary between
the Palaeolithic and Neolithic Periods.
Even supposing that the Deperet system is substantially
sound, it is not easy to disentangle the complications which
arise from such irregularities of earth movement. In my attempt
to apply the Deperet divisions to our English series I have
therefore considered the evidences of relative dating, and not
those of relative level,although the latter is the basis of the system.
My attempt is shown in the table of classifications, but I cannot
feel that it is very satisfactory, while Deperet's own attempt,
and also the modification of Osborne and Reeds (3) (both being
based on relative level) have the effect of throwing the relative
date of our English deposits out of their true sequence. They
have not allowed for the effects of differential earth movement.
THE MOUSTERIAN ICE AGE.
Among all the host of students of the Ice Age, there are very
few who doubt the fact that the Mousterian epoch coincided with
a glacial episode.
James Geikie and Penck placed the Rissian glaciation in the
Mousterian, and this view was accepted by Dechelette. Boule,
however, held that this should have been the Wurmian glaciation,
and if " orthodox " be a proper term to apply to scientific theory,
this interpretation must be given that honour at the present day.
Hoist (4) (vide table of classifications) placed his main