a circular bailey, c.25m in diameter, with a peripheral motte, c.8m high and c.5m across at the top. The bailey is surrounded by substantial banks rising c.2m above the present inner ground level and c.5m above the outer ditch. Entrance to the bailey is via a defensive approach on the north west side that survives as an earthwork between the bailey rampart and the motte. A similar but smaller feature can be seen on the south side. Situated on the north bank of the River Don, the site commands the ancient ford at Strafforth Sands. In the 11th century it was a manor of Roger de Busli, lord of Tickhill. Writing in the 17th century, Dodsworth mentions "Mexborough, where hath once been a castle", suggesting the stone visible in the top of the motte is part of the foundations of a stone tower. (Scheduling Report)
Medieval motte and bailey castle, damaged by landscape gardening. The truncated motte is 52 feet high above the ditch and the bailey is surrounded by a bank 6 feet in height. The ditch surrounding the motte and bailey is 50 feet wide. Although mutilated most of these features were present during field investigation in 1965 and are visible as earthworks on air photographs. (PastScape)
The modern town of Mexborough gives little hint of the medieval layout. The parish church, now well east from the modern town centre was actual on the west of the medieval village with the castle some distance from the village. A footbridge across the river Don marks the approximate site of a ford/ferry crossing but another ford was sited more directly beneath the castle. The castle is not on the highest land (The adjacent primary school is on higher land) but on the false crest of river escarpment making the site visible from the old road and the River Don coming from Conisborough (The old does seem to have followed the line of the modern A6023, along the crest top, and not run closer to the river below the castle as previously stated in this Gatehouse
record). However, the castle is not well visible nor has particularly good views of the land to the north or the river to the west and its view of and from the parish church and village centre is not impressive, being partly shielded.
One of these fords may have been the ford of Strafforth Sands from which the large Wapentake of Strafford took its name, although that may also have been the ford below Conisborough (or indeed the ford may have moved over time according to the changes in the course and flow of the Don). The building of a large canal and the general reduction in river flows because of water extraction mean the modern river Don, whilst still sizeable, is but a shadow of the medieval river which was certainly navigable to Mexborough and probably (with some haulage over shallows) all the way to Sheffield. It has been suggested it may even have been tidal as far as Mexborough. Regardless fording points will have been few and of great importance for policing, administration, taxation and national defence (the practice of Danish 'Viking' raiding and invasion fleets sailing well up tidal rivers ended in 1066 but the threat continued into the C12 - although it is difficult to see, in regards to the Don, how any such fleet would have got past Doncaster).
Mexborough is located towards the north eastern end of the northern branch of the dyke known as the Roman Ridge or Roman Rig. The other end of this north branch is marked by the motte at Kimberworth
. This dyke is now considered to be a pre-Roman Conquest boundary possibly with some use as a defensive dyke during the Roman conquest (see Brigantes Nation
mainly based on Boldrini, Nicholas, 1999, 'Creating space: a re-examination of the Roman Ridge' Transactions of the Hunter Archaeological Society
Vol. 20 pp. 2430) however it may have had some reuse in the early Saxon period, certainly as a visual boundary marker. It is, therefore, possible that either or both mottes are based on Saxon sites built to mark this boundary in some way (The residence of a thegn charged with training and mustering the local fyrd?
While the organisation of the late Saxon Wessex fyrd
is quite well known that of earlier Northumbria and Mercia is less well understood).
The large Domesday manor of Mechesburg
reduced in value from £6 in 1066 to £2 in 1086. It had been held by Ulfkil Brother of Algar
Although Ulfkil still held some Nottinghamshire manors in 1086 Mexborough was part of the large holdings of Roger de Busli
and specifically part of his honour of Tickhill. However it is quite unlikely Roger resided at Mexborough which was, presumably, the residence of a knight, or two, in his service. There is some suggestion from the later tenurial history that there were actually two holdings in the manor as there seem to have been pre-Conquest. 'Roger of Montbegon, with his corpse, gave all his lands in Mexburgh,
as well in demesnes as what was held by the freeholders, as by his villans; with the medieties of the Church' (Dugdale
). Roger de Montbeggon
died 1226. Presumably Roger came to have a holding in Mexborough via his mother Maud, a daughter of Adam fitz Swain (the founder of Monk Bretton Priory) but this might suggest the manor was being administered by a bailiff rather than a tenant with any hereditary rights.
The later medieval manor house of Mexborough
was beside the church, although, if there were two parts to the manor, it may be the site of an earlier manor house.