This description adds to what I have before stated, that there were two western towers, though these were not really erected till later, but were, nevertheless, in all probability a part of the first design. It tells us also of the monastic buildings. Of the scale of this first Anglo-Norman church we have some indirect means of judging. In the first place, it is unlikely that a church of royal foundation, built in juxtaposition with the palace, and intended as the burial-place of its founder, built also in substitution for a pilgrimage which he had vowed to make, should be other than of similar scale to the great churches erected at the time in the country whence he borrowed his architecture. In confirmation of this we have several evidences, not necessary here to state, that it differed but little in scale from the present church; indeed, had it been otherwise, the succeeding historians would hardly have spoken of it in the terms which they make use of. As to its architectural character, we have little to guide us. We have the extensive substructure of the dormitory and the lower part of the refectory. From these we find that the offices were of the plainest variety of Norman; indeed, the pillars of the first-named structure are of the very extreme of massive simplicity, and the shafts of the refectory arcading have cushion capitals of the most normal type. We have recently discovered, beneath the pavement of the altar space, the bases of two of the great piers of the Sanctuary: from which we find that they were clustered, not unlike those at St. Stephen’s at Caen. The bases consist of a double hollow, precisely like one from that church. The work is by no means so rough as that common in early Norman buildings; a circumstance which I have noticed in several pre-conquestal works. Having noticed this one building in which Norman architecture was used in England before the Conquest, I will mention one or two instances of Anglo-Saxon architecture being used subsequently to that event.