face in the jamb, of greater or less width; and that either the wall was thinner than the present one, its interior face occupying the same position, or else that it stood back altogether, as in the front of Peterborough, forming a magnificent porch or atrium. I confess I wish we could prove this latter arrangement to have been the true one, for I am convinced no other could have given a grander effect to the arch, but I have not yet been able to discover in the masonry within the church any indications which might ratify the conclusion. The masses however extend upwards of thirteen feet from the west wall to the first pier-arch, which is narrower than the others in the nave; and the circumstance that no remains exist of a Norman door in the west wall (a feature generally untouched even when later windows are inserted) leads us to suspect that the whole wall belongs to a period subsequent to the Norman.
Some rough masonry seen in the clerestory wall between the vaulting and external roof of the aisle, gives reason to suppose that the original Norman design under Fitz-Hamon (in the reign of Henry I.), included two western towers, which probably were never brought near to completion; the present turrets might be the work of his successor, Robert, earl of Gloucester. The same also appears to be the case with the upper part of the central tower, which differs not only in architectural character, but in masonry and even material, from the lower. This lower stage is perfectly plain externally, with two simple round-headed windows on each face; within, it is ornamented by an arcade enclosing a narrow gallery, and by other arches engaged in the wall, clearly intended to be seen from the body of the church, from which it is now excluded by a vaulting in one of the later styles. The interior of this stage of the tower is of regular smooth masonry, though somewhat wide-jointed, and of a material very similar to Caen stone. The upper part of the tower, which is much enriched externally, and exhibits a range of intersecting arches, is perfectly plain in its interior; which is faced with red sandstone, of a masonry much less highly finished than that below, in fact, giving no indication that it was intended to be open to the church. From this I cannot help thinking that Fitz-Hamon's structure was completed, in execution at least, if not in design, with a central tower consisting only of the present basement-story, and having much
