ON EXCAVATIONS ON ROMAN SITES IN BRITAIN. 233 
season’s excavation work has been much stimulated by the invaluable 
co-operation of one of the Research Committees of Section H of the 
British Association, as well as helped bya grant of 151. from the same 
source. Professor R. C. Bosanquet, the Secretary of this Committee, 
spent five days with us upon the site. 
Attention was first directed to the interior area at the north end, 
and to the artificial defences and an entrance near that end; subse- 
quently the defences to the south and south-east were investigated, 
and finally further examination was made of the south-east entrance. 
Last year evidence was obtained of three occupations of the hill fort 
in a section of three superincumbent roadways in this entrance. It 
was shown by relics unearthed upon the topmost roadway that the 
latest of these occupations was during the fourth century a.p. Many 
similar relics (of which photographs were exhibited) were found 
also in the interior area of the stronghold, proving that portions of the 
hilltop at any rate were inhabited by a primitive-living native resident 
population at that time. 
During the present summer excavations have revealed relics, in 
the form of pottery, coins, &c., belonging to the same period at the 
northern end of the hill fort and elsewhere. It was thought at first 
that the fourth-century occupation of the hill-top might have been 
partial only, but identical remains have now been found within the 
stronghold at the south-east, the south, the south-west, and the 
north, proving an occupation of practically the entire site by a large 
number of people, who, besides possessing implements and utensils 
of home manufacture, used Roman pottery and a Roman currency. 
All these fourth-century relics were found very near to the present 
surface, being covered by one to one and a half feet only of vegetable 
humus. 
Last year a plan showed the fourth-century roadway in the south- 
east entrance as far as excavated. It was a passage with roughly 
built side walls in dry masonry, thirty-eight feet long, cut through 
wide-spreading ruins; it has since been found that it was closed by 
four gates set up at intervals within its course, of which the 
holes for the wooden gate-posts, and some charred wood fragments 
found in a few of them alone survive. Photographs exhibited showed 
that portions of the side walls of this passage were built upon previous 
ruins. Further excavations this season have revealed two guardhouses 
here, one on each side within the entrance; these also are constructed 
amid ruins, their sites being dug out of the fallen débris of earlier 
guardhouses. 
Work during 1912 showed that the inner ditch at the south side 
of the stronghold was filled with the ruin of a wall which previously 
stood on the rampart above. This year’s investigations have shown 
that apparently the whole length of this ditch was so filled as well as 
a similar one at the north end. Cuttings were made across the second 
ditch on the south-east, south, and south-west sides, and it also was 
found to be more or less filled with stony débris in the neighbourhood 
of the entrance and on the south-west side. It was further discovered 
this year that sometimes the first and sometimes the second of these 
