July 7, 2001

Locus 2016 (center of trench ~ 2m x 1.5m) transformed into a wall today as we cleaned it for photography. A single row of larger stones was uncovered on both the W and E edges, with no cobbles outlying, establishing a border to the wall (excavation work was also done in L2021, along the E baulk, ~ 4m x 1.5m, to authenticate this assertion, and aside from a few unassociated, scattered pebbles, there was nothing to suggest the architecture of L2016 extended past the wall edge). There may be a similar wall edge along S face, however, there are a couple larger stones which rest beyond the line which would form the wall. My guess is that they're tumble from a second course or similar that has since disappeared. The wall runs SW-NE.

L2020 (NE, ~ 1m x 1m), which I originally thought was part of the same -surface- as L2016, appears to be a severed portion of the wall, though in a much more degraded condition (no edges are clearly visible, and the cobbles are a bit more scattered). I tried to connect the two loci by excavating the area between them (which I dubbed L2024, and is ~ .7m x 1m), but the locus contained only scattered cobbles. The general shape of the wall is maintained, though, and there is little cobble scattering outside of the area the wall, if it continued as per L2016, would occupy. I imagine the condition of L2020 can be attributed to either a) the same disturbance as the severing of L2020 and L2016; or b) its position lower on the slope, with less material and fill accumulated above it.

I also had a portion of L2022 (SE, ~ 1m x 1m) excavated, around the large stones on along the edge of the locus facing into the trench, but they have no discernable connection to L2016. A few centimeters of L2021 (E side, ~4m x 1.5m) were excavated, and turned up only soil and pot sherds, but nothing to suggest the architecture of L2016 continued past its aforementioned edge.

In the final clean-up of the trench, a second layer of stones similar to the edge stones on the E face of the wall turned up immediately below those which are part of L2016, but with a few degrees different orientation, suggesting an earlier wall or foundation which L2016 was built atop.