Remembering that Prior Conrad was renowned for the glorious paintings with which he adorned the choir, mainly built by his predecessor Ernulf, may we not conjecturally attribute the decoration of this chapel to Conrad's time? It was undoubtedly executed after Ernulf's departure from Canterbury (1107), and before A.D. 1174. If work of such merit and magnitude had been undertaken, between 1174 and 1200, Gervase would certainly have mentioned it. Mr. Westlake has expressed an opinion that the decoration of the nave bears a more distinctly Norman, or Norman-English,* character, than that of the sanctuary, or apse, of this chapel.

The designs seem to have consisted of numerous subjects, from Holy Writ, arranged in medallions; of which several of small size, each containing the half-length figure of a saint or angel, were grouped around one of larger diameter. The borders, of conventional patterns, remain upon some of the squareedged members of the groining arches.

The floor of this nave has been raised, nearly two feet, by the accumulation of soil, so that the base of the central column has been covered up. Two window arches, on the south, seem to be those built by Ernulf. Professor Willis describes Ernulf's crypt windows as being "4 feet wide, 6 feet high, having their sills level with the top of the abacus of the groining shafts, and coincident with the earth-table of the wall; they had merely a narrow chamfer on their outer edge."+

In the north-west corner of the little nave, there is a circular stair which led up to St. Anselm's tower,

*The History of Design in Painted Glass, vol. i., part i., p. 39a. + Willis, Architectural Hist. Cant. Cath., p. 74 note.

VOL. XIII.

E

and the chapel of Saints Peter and Paul. Against the northern pier of this nave, we still see a miserable little fireplace, and chimney, which were inserted by the French Protestants when they used this chapel, as the Vestry of their Elders. The joists of a wooden floor, likewise inserted by them, have left their marks upon the soil at the east end, causing it to look like a series of parallel grave-mounds. This flooring, and the wooden panelling with which the French cased the walls, were both removed many years ago.

The east end of this little nave was, during the middle ages, a straight wall, in which towards the south there was a piscina-like niche, beneath a simple pointed arch. The central portion of the wall was very slightly recessed, as if for the reredos of an altar. Northward of it, traces still remain of a bracket for an image, or a lamp. Behind the altar,

there was, low down in the wall, a rectangular aperture (marked AB on the plate), 22 inches high and 18 inches wide, through which persons could with difficulty creep. It admitted those who performed this gymnastic feat into a small apse, perfectly dark, but exquisitely adorned with painting. The apse corresponds in shape, in position, and mainly also in size, with that of the Chapel of the Holy Innocents on the north side of the crypt.

Respecting the mysterious walling up of this apse, there are a few facts upon which we may found some reasonable conjectures.

The recessing of the wall as for the reredos of an altar, and the insertion in it, north and south, of a bracket, and an early arched piscina-niche, prove that the apse was walled up before the period of the Reformation. Upon examining the wall itself, it was

found to be several feet thick, and it became obvious that, ancient as its outer western face undoubtedly was, that facing had been added, long after the apse had been first walled up. The original blocking-wall had been plastered, and painted, before the existing outer facing of stone was added. When this fact was discovered, in 1879, the idea of pulling down the blockingwall was abandoned, because it contains the only existing evidence of date; so that nothing more than a rough doorway, of sufficient size, was opened (or perhaps reopened) through the wall. The top lintel of this doorway is a moulded stone, of the Transition or Early English period. When, then, was the apse originally blocked up? Mr. J. Brent, in his second edition of Canterbury in the Olden Time, p. 283, says, "late in the thirteenth century, in order to substitute a square east end, a wall was built on the chord line of the apse." The arch of the piscina-like niche in the outer face, and the moulded stone in the head of the doorway, now open, might perhaps suggest this date; but, if so, it must be that of a subsequent addition to the wall, not of the original blocking up of this apse. Perhaps, after the interment of the Countess of Athol, a chantry altar may have been dedicated, outside the blocking wall.

The fact which determines the early period at which the apse was first blocked up, is its omission from the minute account, given by Gervase, of every apse and altar in the crypt. Writing in, or about, A.D. 1199, he gives a systematic survey of the whole building, as it appeared in 1174; and, after naming any altar in the choir, he immediately mentions the altar which stood beneath it in the crypt. When, however, he comes to St. Anselm's tower, and mentions the altar there,

of Saints Peter and Paul, in the upper church, he entirely omits to mention the altar-site in the crypt beneath. This is the only similar spot which he does not duly describe. As Professor Willis says, "Saint Gabriel's altar is not mentioned by Gervase, but is known by the ancient painting and inscription which still remain."*

It is certain that Ernulf designed the apse to contain an altar; and it is equally certain that this very beautiful chapel contained the altar of St. Gabriel, in A.D. 1174; why then, in 1199, did Gervase omit to mention it? The only possible reply must be, that, for some cogent reason, the monks had already shut up the apse, and desired to keep its existence secret. One side of the necessary window-arch of Ernulf can still be traced, in the north part of the east wall of the apse, but it has been admirably blocked up, with wellfinished Norman masonry, so that upon the exterior of the building there is no sign whatever of any window or chapel, in the apse. It is evident that the monks, in blocking this apse, sought the greatest possible degree of secrecy. The apse of The Holy Innocents' Chapel was precisely similar in character; but it was not so private. Being close to the Priory, and near the monks' entrance to the north transept of the crypt, it would be passed continually, from hour to hour, by monks and by pilgrims. On the other hand, St. Gabriel's chapel was so entirely out of the way, that in St. Anselm's tower, above it, the monks subsequently constructed the secret" Watchers' Chamber;" from which they kept constant watch and ward over Becket's Shrine, in the Trinity Chapel, above. Even the exterior of St. Gabriel's apse was at the obscurest

Architectural History of Canterbury Cathedral, p. 39, note m.