The east side of OA25 is relatively extensively investigated within Excavation Area N. The eastern boundary with OA24 is assumed to persist in some form in Period 3, some early Roman features being either positioned alongside or on its line (Figure 99).
The lack of early Roman encroachment upon the footprint of Building 32 is notable, and it is possible that it survives through much of Period 3. Mid-2nd century well 8989 is construed to cut the south wall, presumably marking the building's demise (unless a modified and shortened building plan is accepted i.e. not including the post-built southernmost room). In contrast to the west side of the enclosure, the east appears to be occupied by widespread pitting, some of which clearly has a more specialised (primary) function than that of mere rubbish disposal.
Feature 25221 is a very large, sub-rectangular cut on the northern edge of Excavation Area P and extending beyond the limit of investigation. As exposed, it is 6m wide and in excess of 12m long and would seem to be a single pit some 0.8m deep. It may reasonably be speculated to extend northwards as far as the edge of Road/track 3, making it some 27m in total length. Its edges and sides, where recorded as excavation segment 8912, are fairly irregular and of variable slope and contain a sequence of gravelly silt fills that display at least two episodes of re-cutting or partial cleaning - 19176 and 19177 (Figure 121). These contain relatively large assemblages of pottery and tile, though perhaps not when considered in proportion to their volumes. Although created in the late 1st century AD, deposition continues as late as the 3rd century.
Interpretation as a gravel quarry is unconvincing, as is its use purely as a rubbish pit in the absence of widely varied artefactual assemblages and in particular animal bone which should survive well in such a large feature, if present. The positioning of this feature over and following the line of the Period 2B plot boundary ditch between OAs 25 and 24 (25178) is surely significant. Regardless of its specific function, 'pit' 25221 seems to be deliberately placed upon, and to perpetuate, this boundary. In this respect, as well as in dimensions, it is reminiscent of the Period 5 pit 25212 in Northern Zone OA31.
Pit 8540 is a small, 1.5m square, pit with near-vertical sides and flat base. It contains a lining of clean orangey-brown clay in which vestiges of a wooden structure are visible (Figure 122). Four stake-holes (8619, 8621, 8623, 8625), each located in a corner, mark the positions of uprights to support an inner lining of planks that has decayed. Nails, used in its construction, survive more-or-less in situ where they have been driven into the clay outer lining through the uprights. The clay also retains impressions of grooves or steps, both near the bottom to support a wooden base and at the top of the feature to accommodate a collar or perhaps even a lid. Surviving to a depth of 0.5m, 8540 is cut into the fills of large early Roman rubbish pit 8524 (see below). This may have been deliberate, perhaps for ease of excavation or to exploit the damper, cooler, insulated conditions afforded by the silt fills as opposed to the surrounding natural gravel.
A deposit of find-free fine silt (8570) lies below the level of the lower groove and presumably accumulates while the wooden base is in place. However, its artefact content gives no hint as to the primary function of this lined pit. The remainder of the clay-lined pit is backfilled with rubbishy silts that contain pottery, small amounts of animal bone and tile, and a single (?residual) loomweight fragment. There is thus no clear evidence of function, though it is likely the structure is intended to retain water. A use as a live fish or shellfish store is possible. A very similar feature at Gorhambury, differing only in being twice as long to the same width, is interpreted as a two-person privy (Structure 44; Neal et al. 1990, 68-9, figs 93-4). The Elms Farm example could perhaps be a single-seater version, though quite why this would require such an elaborate lining, surely preventing effective drainage, is unknown.
Structure 34 is a sub-rectangular building located toward the southern end of OA25. Post-holes and slots (25175, segs 8653, 8673, 8682, 8684, 8686, 8695) define the eastern foundations of a building c. 5m wide (Figure 123). Its length is unknown, being obscured by inter-cutting features to the west. However, a possible westward continuation of its south wall is suggested by a line of unexcavated post-holes recorded on the pre-excavation plan, and indicates a length in excess of 8m. The 2.2m gap between gully 25175 and post-hole 8684 probably constitutes an entrance at the east end of the structure. Only gully segment 8653 contains artefactual material, nine small sherds of ?early Roman pottery, and is cut by 2nd century AD pit 8515. No internal surfaces or features are discerned.
Some of the surrounding scatter of small pits (Group 659) may be associated with the use of this building, along with 'storage' pit 8540 (Group 657) positioned little more than 2m to its east and very convenient to the entrance.
Occupying a central position toward the perceived rear of OA25, and cut by lined pit 8540, pit 8524 (Group 656) is a substantial sub-oval cut some 6m x 3.7m and c. 1.0m deep. Its finds assemblage largely comprises early Roman pottery and an iron knife blade (SF7720).
Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.