NEOLITHIC AND LATER PREHISTORIC LANDSCAPE OF THE AVON VALLEY 223 
flakes and a core were recovered from the fill, and it 
is probably prehistoric. 
Site 4 lay to the east of Site 3 and included a 
number of features spread over about 500m. The 
most westerly was 174, a small irregular depression. 
This was roughly oval in plan, with a flat base and it 
produced a large amount of burnt flint (1.79kg), 11 
fragments of animal bone and three flint flakes. The 
fill was homogeneous in character, indicating that 
the feature had been backfilled in a single episode. 
LATER FEATURES 
At the reservoir, an inhumation of Romano-British 
or later date was found lying within a ditched 
enclosure (188) (Figure 2, Site 1). The human 
remains were those of an immature individual 
(Jenkins, pers. comm.), about 12 years of age, 
within a grave. A water pipe trench had previously 
destroyed around 75% of the grave. The remainder 
was excavated, revealing both legs and feet below 
the knees. A quantity of disarticulated bone was 
also recovered, and only the arms and a tibia remain 
missing. The grave was aligned west-east, and was 
filled with very loose vacuous chalk rubble. The in 
situ legs and feet were surrounded by square-shank 
nails, which appear to have been part of a coffin. No 
datable material was recovered from the grave, but 
the west-east style of burial and the presence of a 
coffin suggests a Roman or later date. 
A ditch (186), situated to the east of the 
roundabout and running north-south, was possibly 
of later prehistoric date (Figure 2, Site 3). Post- 
medieval field boundaries and wheel-ruts were also 
encountered (Figure 2, features 159, 161, 167, 170 
and 172). 
UNDATED FEATURE 
Ditch 146 was situated at the junction between the 
access road and the A3028 (Figure 2, Site 1). There 
was no time available for this feature to be 
excavated because of its position on the access 
road. It was, however, sealed, and thus preserved, 
below a layer of ‘terrain’ and hard-core. The 
‘feature was 1.7m wide, and ran across the line of 
the access road on a west-south-west to east-north- 
east axis. The ditch was filled with a light 
yellowish-brown fine silty clay, a much finer 
deposit than the fills of other features observed in 
this area, and appears to have been washed into the 
ditch by successive periods of rain. No finds were 
recovered from the top of this fill. 
POTTERY 
by Elaine L. Morris and Rosamund M.f. 
Cleal 
Twenty sherds from a single Grooved Ware vessel 
were recovered from pit 165 (Site 2; Figures 2 and 
4). The form is not reconstructable, but the three 
conjoining sherds appear to belong to the base and 
lower part of the vessel. The fabric is soft and 
contains sparse quartz sand (0.5mm), rare iron 
oxides (small reddish grains responding to a 
magnet), rare calcareous fragments (shell or chalk; 
the fragments are too small to identify), and some 
grog. The grog is difficult to distinguish from the 
matrix. The colour is pale brown (exterior), dark 
grey-brown (interior) and the core black. The 
decoration consists of incised lines and jabbed 
impressions, with at least one wavy applied cordon. 
It is not possible to reconstruct the arrangement of 
the decoration, but areas of jabbed impressions do 
occur on vessels with zones of incision, although 
the impressed decoration tends to be confined to 
the upper body, as for instance on P77 and P229 at 
Durrington Walls (Longworth 1971, figs 38 and 
49). The wavy cordon is also paralleled at 
Durrington Walls, where there is one sherd with a 
smooth wavy cordon (fig. 36, P58) and several with 
wavy rusticated cordons (fig. 44, P162-167). The 
example from pit 165 seems more similar to the latter 
Fig. 4 Grooved Ware from Site 2, pit 165 
