Five definite wells and a possible sixth example were identified. Three wells were located within the northern half of the site, the possible well within the central part of the complex and the last two wells to the south, within the large enclosures divided by ditches.

One well was situated within the possible structure and had a construction cut that measured 2.25m across, though this narrowed to c 1.60–1.70m between the stone lining. At least five courses of the stone lining remained in situ, constructed from roughly hewn stones. The upper part of the lining appeared to have been cut away, presumably representing an episode of stone robbing. The surviving lining was only 0.5m deep, below which the sides of the well were formed by the natural clay into which the shaft had been dug. The well exhibited evidence for two distinct phases of use, the earlier of which was 1.5m deep. A homogenous main fill of silty clay contained pottery dating to 270–400 and was overlain by a second fill of clay silt containing charcoal, residual 2nd/3rd-century pottery and animal bones, suggesting that it was a dump of material. After this dumping episode, the well was recut and appears to have been cleaned out. This cut was filled with another homogenous clay silt containing more residual 2nd-century pottery. Possible remains of a superstructure over the well are indicated by four postholes. These cut clay packing indicating that they were associated with the recut phase of the feature. They encircled the well leaving a gap, possibly for access, on the south-west side.

The construction cut of a second well measured over 2.3m across and had straight, vertical sides. The stone lining was faced on the internal surface and was two courses thick. The internal diameter measured 0.8m and the well was excavated to a depth of c 3.2m. Most of the shaft was filled by a single deposit of dark grey silt that contained sizable quantities of pottery and animal bones indicating that it was backfilled with material that included domestic refuse. The pottery from this fill was dated to 250–400. The feature was cut on its south side by a furrow.

The cut for the third well measured 2.3m across. The lining comprised several uneven courses of roughly faced limestone blocks. The lining was notably more poorly constructed than in the second well. The western side of the stone lining was robbed out of the well, presumably during a period is disuse. The feature was later recut with the stone lining on the opposite side remaining in situ. The shaft was filled by a single, soft, grey-blue clay silt containing occasional large stones and a fairly sizable quantity of pottery, animal bones and tile, suggesting waste disposal. The pottery from this fill comprised 14 sherds (288g) dated to c AD 270–400 based on the presence of bowls or dishes in Nene Valley colour-coated ware, including a copy of samian form Drag. 38, and a dropped flange bowl in sandy reduced ware. A pit was dug into the western side of the feature but was undated.

The fourth well was located close to a ditch. The construction cut was 3m wide and had an unbonded limestone lining constructed using roughly finished blocks. The stone lining was excavated to a depth of 1.22m before further excavation was prevented by ingress of groundwater. The clay packing behind the stone lining contained fragments from a Drag. 31R dish in Central Gaulish samian ware, a hook flanged mortarium in Mancetter-Hartshill white ware and a flanged dish or bowl in medium sandy reduced ware, indicating that it was constructed no earlier than the second half of the 2nd century. The well shaft was filled with a soft, waterlogged clay that produced some poorly dated Roman pottery. The top of the well had been disturbed and early medieval inhumation burial inserted into the top of the feature.

The fifth well was very disturbed by later robbing. It had a wide construction cut measuring 3.2m across and a thick clay packing around the sides. This was cut into by a large robber trench, presumably dug to remove the stone lining, and backfilled with limestone rubble, a mixture of pottery (dated to 230–350), animal bones and tile. This fill was only excavated to a depth of 0.58m, which was sufficient to define the top of the stone lining. The well was cut on the western side by a pit, which was then cut by another pit in the centre of the well. Both pits contained later Roman pottery.

The sixth feature may have also been a well owing to its size, measuring 2.8m across and reaching at least 1m deep. The feature was cut through the centre by a ditch and contained a single fill that produced later 2nd-century pottery.

Martyn Allen and Lauren McIntyre, 2019, Chapel Lane, Bingham, Nottinghamshire: Archaeological Excavation Report, Page 25-26 (Unpublished document). SNT5816.