PILE-DWELLING SITE AT SKITT'S HILL, BRAINTREE. 145
growth, the land cultivation and the baring of the virgin soil. Bearing this
consideration in mind, we may say that the pre-historic accumulation was
perhaps ten times slower than the Roman, and that of recent times much more
rapid than the Roman and Mediaeval."5
With the section as now revealed, this view must be consider-
ably modified. Instead of the relic-bed resting directly on the
river-ballast, we have beneath it an alluvial accumulation of
7-8 feet, or a total depth from the surface of 17-18 feet. Mr.
Kenworthy has given us the depth of 4 feet 6 inches from the
surface as the lower horizon at which Roman objects occur; we
now have 13 feet of deposition below this representing pre-
historic accumulation.
As the silting-up of our present river-system is generally
considered to be coeval with the commencement of the Neolithic
period, the assignment of the Skitt's Hill pile-structures to this
era was, perhaps, incontrovertible so long as their foundations
were said to rest close on the river base, and Mr. Kenworthy
might well suppose that the silting-up of the river was so much
slower in pre-historic times, if after continuous occupation
through the long period of the Neolithic and Bronze ages only
about three feet of filling had been accumulated, while above
was about 6 feet 6 inches, representing the comparatively brief
period of historic times.
The brick-earth provides very little reliable evidence as to
period, owing to its being the result of rain-wash. It may conse-
quently contain objects derived from higher levels, which would
not occur in any regular order of superposition. Romano-
British pottery extends more or less throughout this deposit.
Although some of the pottery may be of the class now recognised
as late-Celtic, it is impossible to decide this from the fragments
that have come to hand. Many of the forms of late-Celtic
pottery survived into the Romano-British period, and are
commonly found in association with the more purely Roman
shapes. It is only when late-Celtic pottery is found quite
distinct from Roman relics, that it can be taken as any indication
of an earlier age.
In the absence of positive evidence, therefore, I have thought
it better to omit any portion of the rain-wash as being late-Celtic,
and simply to divide the whole filling of the stream broadly into
historic (including Romano-British) and pre-historic.
5 Ibid. p. 99.