510 Inaugural Address by the President of the Society, 
Traverse was thrown up for the purpose of continuing the entrench- 
ment to the westward, the soil did contain Roman remains, whilst, 
at the time when the older portion of the dyke was thrown up, the 
same ground did not contain Roman remains. 
Two hundred and sixty feet of rampart, in all, was dug on the 
south-east of the epaulement without finding anything Roman 
except some dubious pieces of pottery, quite near the surface. The 
bulk of the pottery was of a kind that might be attributed to the 
British as well as the Roman Age. This goes a long way towards 
proving that the dyke to the south-east of the epaulement was 
earlier, and that the extension of it to the north-west was made in 
Roman or post-Roman times, but it is not conclusive. This spot is 
more distant from the settlement than Sections 1 and2. Whatever 
kind of pottery exists in the soil will be thrown up into the rampart, 
and at whatever period a rampart may be made it will disclose only 
such kinds of pottery as the soil contained, or such as might have 
been accidentally dropped into it during its construction. The 
absence of: Roman pottery is, consequently, no proof that a rampart 
is earlier than the Roman times, though it may leave the question 
of date open. 
Trenches were dug in the combe at the end of the epaulement, to 
ascertain whether it had ever extended further and been destroyed 
by cultivation, but the end of the ditch was found at a distance of 
272ft. from the spot where it leaves the Main Dyke, showing that 
it never extended over the hill, but must have been merely a short 
turn of the rampart to cover and protect the exposed flank, at this 
time probably also protected by a dense growth of trees and under- 
wood. 
The question of the age of the right flank, right centre, and a 
considerable part of the left centre of Bokerley Dyke as far as the 
epaulement, must be left for future investigation, before it can be 
determined with the same certainty that we can now speak of the 
left flank. I have only to say, however, that as the whole character 
of the extensions coincides with that of the main portion of the 
entrenchment, except in being of slightly less relief, there is a 
probability of the latter being found to have been constructed by 
- 
ee 
