But running just inside it, barely visible beneath the grass, is an older ring of earth that predates the wall by an unknown stretch of time, and which tells a more complicated story about how this ground came to be set apart.

When the Ordnance Survey mapped the area in 1840, the graveyard appeared on the six-inch sheet as an unenclosed space, marked only with a dashed line. By the time the revised edition was produced in 1904, a proper stone wall had been built. What survives beneath it, however, is the ghost of an earlier enclosure: a broad, low earthen bank, roughly 2.4 metres wide at its base and less than a metre high externally, that traces the northern and eastern edges of the site. Along the western side it survives as a scarp rather than a raised bank. In the north-east quadrant it has been partially destroyed by a nineteenth-century vault and burial plot, and along the southern edge it disappears entirely, though the ground falls sharply there, which may itself have served as a natural boundary. The graveyard measures roughly 41 metres east to west and 35 metres north to south. Among the headstones, which are predominantly eighteenth and nineteenth century, there are two older survivals located to the south of the church: a late medieval graveslab and an armorial plaque, the latter carved with a coat of arms, suggesting that someone of local consequence was commemorated here long before the existing wall was raised. The entrance lies on the western side, facing the road.

Visitors approaching from the west will find the entrance directly off the roadside. The earthen bank is most legible along the northern interior edge, where it runs parallel to the stone wall; the eye adjusts once you know to look for a slight rise in the ground rather than anything dramatic. The late medieval graveslab and armorial plaque, set to the south of the church, are worth seeking out for the contrast they offer with the more familiar slate headstones that dominate the rest of the space.