With the removal of the precinct wall, the religious focus is now open to the east and the clear distinction between OA23 and OA45 must now be blurred - at least at their interface. The temple is presumed to survive as a standing, though not necessarily functioning, building into the 5th century. The survival of Building 63 is unclear.
The imposition of a substantial rectangular building (Building 64) across the former precinct wall line and partially in front of the temple marks a crucial departure in the development of this complex. This is perhaps the clearest indication that, even if the temple is still in use, it is no longer the principal focus of OA23. No other features are in obvious association with this new building, which is assumed to function at least to the very end of the Roman period, perhaps even beyond. Nor is there anything of overtly Early Saxon date within OA23. It is conjectured that the temple structure persists as a revered or avoided ruin in the 5th century.
It is possible that Building 63 passes out of use relatively quickly. Within its interior, silty sand-clay deposit 5416 (Group 453) accumulates, which might suggest the building has lost its roof. This deposit seals infilled pit 5450, but is cut by a second internal pit (5400 Group 454) - a partial re-cut of the first. This also contains lead items (SFs 2069, 1004) along with a likely capping deposit of tile.
Finally, the building remains are covered with apparent abandonment and destruction deposits (5383, 5386, 5387, 5441 Group 455) that includes quantities of tile, stone rubble and pottery. Further lead waste (SFs 1919, 2030, 8386) is also present, along with copper-alloy items and Late Roman coins in the overlying cleaning layer (5383).
Although already mentioned, the dark sandy silts (5148, 5153, 5159, 5227, 5228, 5275, 5305, 5306, 5307, 5401, 5409, 5413, 5414, 5453, 5543, 5580, 5605 Group 457) that overlie the latest Road 4 surfaces also overlie various features along the northern edge of the temple precinct.
Although these strata were treated as essentially cleaning layers in the field, they do seem to represent real archaeological accumulations. All contain considerable quantities of finds, and all date to the late 4th century or later. The nature of the finds assemblages suggests that these deposits are principally the ill-defined upper portions of the roadside pits and ditches that they overlie - although an element of road accumulation and road-wash is not discounted. Having identified an element of structured deposition in the features below (see Central zone, Open Area 23),it is unsurprising that these deposits contain similar ranges of finds and include some 126 mid- to Late Roman coins. As such, these deposits should be interpreted in close conjunction with the late pits of this area, though acknowledging that a significant degree of contamination from the Period 5 roadside pits is also very likely.
Building 64 is a timber structure that is constructed, perhaps deliberately, over the line of the former eastern temple precinct wall following its removal. Lines of closely set post-holes (18873, 18879, 18889, 18952, 18953, 18992, 21006, 21008, 21010, 21013, 21504, 21506, 21509, 21511, 21525, 21527, 21529, 21531, 21554, 21580, 21583, 21589, 21721, 21730, 21732, 21738, 21748, 21783, 21818, 21820, 21884, 21888, 21890, 21892, 21905, 21923, 21936, 21947, 22010 Group 456) of generally uniform size define the walls of a rectangular structure at least 12m long and 7m wide (Figure 207). A smaller rectangular room, measuring approximately 4m by 5m, extends off its western side. The different construction of this room, with a slot foundation, suggests that it may be an addition to the main building. Whether the full northwards extent of the building is established is unknown, but it is likely that it does not impede access from the east to the temple itself. An internal partition of the main room is postulated.
While the post-holes of the main part of the building are largely uniform, those at the corners (such as 21884 and 21529), and occasional examples along the walls (e.g. 21527), are larger and contain a gravelly clay packing. These may house the main supporting timbers and help define the extent of the building. The slot foundations of the additional room are somewhat irregular. However, slot 21013 contains post-settings and slot 21006/18892 terminates in further integral post-holes, one of which (18953) at the south end of the slot seems to be replaced by a tile, flint and chalk post-pad (18924) bonded with opus signinum. This may hint that the interior, of at least the added room, is also surfaced with this material. Slot 18892 also contains large sandstone boulder 18963 and two smaller similar fragments. These are positioned at the corner of the foundation in order to give it a solid footing. The smaller room clearly contains a plain tessellated floor, which survives as in situ fragments (18923) sealing the offset south wall foundation.
Silt layers 18923 and 21514, and overlying cleaning layer 21500 (Group 476), may be the remains of accumulated use deposits within the main building. These layers and many of the structural features contain quantities of tesserae, but deposit 21500 is noteworthy for the twenty-one Late Roman coins collected from it, which may well derive from the occupation of this building. Small 'post-hole' 21745, roughly in the centre of the building interior, contains a complete tazza bowl and is also tentatively associated with its occupation - perhaps a foundation offering placed beneath its floor.
Building 64 is the clearest demonstration of latest Roman activity within the central zone of the site. It straddles the remains of the demolished precinct wall and represents a major departure from the prevailing layout of this area, which lasted for almost 400 years. At the same time, a degree of conformity is maintained in the alignment of the building, which follows that of the underlying wall. It is likely that Building 64 does not extend significantly further and that it respects the access to the temple. This is not to say that the building is necessarily associated with what survived of this religious complex; it could as easily be a secular structure. However, it is tempting to see the late coin assemblage from deposit 21500 as being similar to that in well/pool 22210 (Group 987). Together with the tazza deposit in its floor, a shared religious association with the temple could be construed. The possible function of Building 64 as a religious building (perhaps early Christian) is further explored in Atkinson and Preston (2015, Chapter 6).
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.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.