When valour bow’d before the rood and book,

And kneeling knighthood served a Prelate lord.

The window recesses all through the building are mostly in the Norman style, and therefore in almost every case much older than the windows which they contain.

The three ends of the curtain abutting on the keep are of the same workmanship with, and bond into it, and are about 30 feet high. That proceeding northwards is capped at the north-east angle by a stout bastion, with a salient angle of 110 degrees, two faces of 17 feet, and shoulders of 4 feet. This is evidently a rebuilding after the introduction of artillery, and probably the work of Sir George Bowes. This curtain is continued 30 feet further along the river face, and is thence broken down. The kitchen seems to have been in the angle, the hall next to it, and then the chapel, all built against the north curtain. The curtain now standing corresponds closely to what the survey of 1515 calls the long high wall from the Dongeon to the north-west end of the kitchen (44 yards long and 30 feet high, contremured, so as to be 28 feet thick). We are told that in 1551 the chapel was 30 feet by 18 feet, with walls 8 feet thick, and with a crypt below capable of stabling 20 horses, and a “closet” above, and that the battlements above this closet and of the long wall were of the same height, and so extended from the north-east angle of the keep round to its south-west angle. In Sir George Bowes’s very able report upon the castle, he advises strengthening this wall by filling the hall and other buildings with earth, and forming a hall in the first floor of the keep, which seems then to have been much such a ruin as at present. Passing to the south-west angle of the keep, whence springs the southern curtain of this inner ward, also 30 feet high and very thick, this is continued 50 feet, and then broken by a nearly rectangular bastion tower of 40 feet projection, and 30 feet breadth, the “little Bulwark” of 1551;—a Decorated insertion to give a flanking defence where it was much needed. In the rear of this tower, which probably was of the nature of a bastion, that is not higher than the curtain, are remains of buildings. Beyond it, a high bank of earth and rubbish marks the line, but conceals the remains of the rest of the curtain. In this bank, towards the west, a gap marks the position of the gatehouse. The well remains near the north-east corner of the ward, and indicates the general position of the kitchen.

The Outer Ward, or that part of the castle outside the ditch of the inner ward, is of a lunated figure, 50 yards at the widest part. This ditch was crossed at each end by the curtain. The lower part of the wall at the east end still remains, and is about 8 feet thick, and 30 feet high, pierced by a Norman arch, probably for a postern, as at Carlisle. The curtain along the northern front is gone. The slope is there very steep, and in Sir George Bowes’s time this was trusted to, and the wall was a mere low breastwork. More to the west, as far as the lower gatehouse, the wall has been rebuilt. It is in parts about 15 feet thick, and 10 feet high inside, and from 15 feet to 20 feet outside. It probably rests in part on a Norman foundation, but the superstructure looks Decorated or later. It was pierced by deep recesses 11 feet broad under a flat arch, splayed to a loop, and intended to flank the approach from the town to the outer gate. Three of these recesses remain, and probably there were two more.

The lower gatehouse is a rectangular block 40 feet long by 20 feet broad, and of 30 feet projection within the curtain. It is pierced by a passage 15 feet wide, reduced at each end and in the centre by gate piers to 12 feet. These piers carried ribs to stiffen the barrel vault of the passage. The arches were round-headed. There was no portcullis. In front of the gate, as at Tickhill, are two projecting walls, between which there was probably a drawbridge. There was an upper floor. This gatehouse is evidently Norman, and no doubt Flambard’s work.

From the gatehouse eastward for about 130 yards the curtain is represented by a high and steep bank of earth and rubbish, which, no doubt, contains its foundations. The ground rises, and the original bank, as well as the curtain upon it, were evidently raised to command the platform opposite and beyond the ditch, which seems to have been the favourite position for besiegers. The curtain recommences, and is continued for 80 yards along the high ground, forming the east end of the southern front. This part of the curtain is very curious, and not a little difficult to understand. Upon it are the remains of one, or perhaps two, polygonal bastions, but in the line of the wall, between them, are six round-headed arches of about 12 feet span, springing from square piers about 3 feet or 4 feet broad. Most of the masonry is so rough that it evidently was intended, as at Southampton Castle, to be covered with earth, though why, while the foundation was excellent, this mode of supporting the curtain should have been employed, is not clear; but the first, that is, the most western, of these arches is of ashlar, and seems to have been a gateway, and is probably the gateway mentioned in the history of the castle as having had its gates unskilfully hung upon gudgeons so placed that they could be lifted off from the outside. Beyond these arches and bastions the curtain makes a sharp turn, and proceeds northwards to 60 yards, to cross the ditch and join the keep.

In front of the lower gate is a small platform, beyond the ditch, and which was, no doubt, stockaded to cover the entrance and enclose the barriers.

The archway described as probably a gate may have been intended to facilitate the entrance of the villagers and their cattle, in the event of a raid. The outer ward was intended by Flambard, it is said, to afford this shelter, and the deep outer ditch was also so employed. In late times a complaint is made, that whereas formerly the castle ditch was, under all circumstances, a place of security for the villagers and their property, now they are attacked and captured by the Scottish rievers under the very walls of the fortress.

The southern and outer ditch is said to have been used as a mill-pool, its mouth being closed by a dam. This ditch is reputed to have been excavated in 1495; but though it may have been then deepened and a dam formed, it must always have been part of the original defence.