10 THE ESSEX NATURALIST.
very surprising, and serve to throw the local river gravels into
a new perspective, with repercussions extending to more distant
areas.
Mr. Kennard remarks that this new, and almost unique,
discovery at Little Oakley finds its nearest comparison with
the West Runton (the earliest) phase of the Cromer Forest Bed,
although it is not identical with it and may be somewhat earlier.
If that should be confirmed it will be the earliest phase of the
Forest Bed group, and in any case it doubtless belongs to the
same river system.
One of the commonest species is a Nematurella that forms a
close link with West Runton, but Mr. Kennard finds that the
whole assemblage is different, although more closely allied to
the earliest phase of the Forest Bed than to any other known
deposit. For example, one species of Unio is characteristic of
West Runton while a different species, not previously recorded
from Britain, occurs in corresponding deposits on the Continent;
both these species are found together at Little Oakley.
When working the Palaeolithic gravel of Upper Dovercourt
I divided the bones into three groups: (1) Crag bone, (2) Derivative
bone that was not Crag, (3) Contemporary bone. These bones
were fragmentary, and unimportant in themselves, but it now
seems probable that the second group came from the neighbouring
"Forest Bed" deposits; this serves to illustrate another trap
in the mixture of dates from the same deposit.
When the Upper Dovercourt pit was extended further to
the south, the rich palaeolithic finds suddenly ended, and there
was no trace whatsoever of any implement or flake. The greater
part of the gravels of the Essex plateau are equally barren.
It is thus clear that the Upper Dovercourt site, which is of very
late Acheulean date, belongs to a channel deposit.
For a number of years I have been watching a large pit on
the floor of the Lea Valley close to Broxbourne station. This
cuts through a group of Holocene alluvial deposits consisting
of peat, marsh clay, shelly sands, etc., in considerable variety.
In these as in many other alluvial deposits there is no sequence
of stratification over the whole area, but a complex of silted-up
channels, one cutting through another: in this case their date
extends from the Boreal and Mesolithic Period to the Mediaeval.