would be obtained through the triforium gallery where two openings show that they were for ingress; on the south side the lowest tier of arches have balls in the hollow of the archivolt similar to the ballflower in application. In two of the bays also, as well as one on the north side, the columns instead of being of three shafts are of one column with reveals carved into lozenges with a raised quatrefoil flower in each lozenge; these are continued round the head.

There is likewise on the south side a fireplace inserted, and there has been a doorway, the shafts having been cut away for the purpose; the former would appear to have been done not very long after the original construction.

It may be well here to allude to an error of importance fallen into by Mr. Britton1 and others in reference to the string courses from which the columns spring. This has been repeatedly described as a seat, for which purpose it is much too narrow, and being also seven or eight feet from the floor, it is obvious that it was altogether above the monks' stalls. There may have been hangings or coats of arms in the recesses, as the mortices seem to show.

The origin of the assertion seems to have been derived from the author of the letterpress of the Beauties of England and Wales,2 and repeated by Britton in his later works -he probably never saw it, or he would at once have discovered that five feet of earth removed would alter the relative levels so much as to render impossible such a supposition. The view in the Beauties of England and Wales,3 shows the entrance buried to within a diameter of the level of the capitals of the shafts.

The great cloister exhibits only bare walls. The string courses show the position of the former roof, by which we see that it was not vaulted; they further show the width of the ambulatory. The outer wall next the garth is entirely destroyed. There were the usual two doorways into the nave, one of which is no longer in existence; the other, which opens under the lower part of the aisle at the western end is tolerably perfect. The cloister curiously diminishes in width toward the west. It is remarkable, that in the abbey of Cluny this part of the buildings has a similar departure from the rectangular plan.

We now proceed to the conventual buildings of the Early English period. Northward of the chapter house in the cloister is the triple groined recess before noticed (B on the plan). It is four feet eight inches in clear depth with the three front openings about eight feet three inches wide, with corresponding recesses at the back. These last exactly match those in the chamber above the crypt of the north transept. I am disposed to regard the arrangement as purely accidental, arising from the increase of the size of the church and the desire to preserve the chapter house, the line of which had to be continued to the former; and with the usual ingenuity of the age to which it belongs, the accident was turned to

1 Architectural Antiq., iv, p. 63.

2 Vol. xiii, p. 200.

3 Ib., p. 196.

account and something useful and beautiful at once produced. It was not the usual position for a lavatory, although, from the absence of any indication of it anywhere else it may have been one; but its most probable use was that of a cupboard where the everyday books were kept, this being the exact position one would point to as complying with the descriptions of Dugdale1 and Viollet le Duc,2 I therefore am inclined to consider it the "Petite Bibliothèque". It has, in almost all parts, what are technically termed straightjoints, which show that all but the main walls was built subsequently to the other parts.

The masonry and the doorways which now partly fill up the original archways are modern, the dressed stones having been re-applied from some other part of the building.

Facing the chapter house was the dormitory, now altogether swept away. The guest house was further to the west; nothing remains of it. From the north end rose the stair which led to the upper floors containing the dormitories and the chamber over the aisle, as well as to the gallery in the triforium, which, as there was no upper floor to the ambulatory of the cloister, will have been used as passage rooms, and the gallery as a means of communication. On the south side of the cloister was the refectory, two walls of which are yet tolerably perfect. It was a lofty room, too lofty, apparently, to have had any room over it, and it was groined in eight bays; it was thirty-two feet wide, by eighty-five feet long. A noble room it must have appeared with its lightly moulded ribs springing from carved corbels, while its lower part was hung, probably, with tapestries, or its walls covered with frescoes. The kitchen, probably, was at R, Plate x.

Amongst the more distant buildings is one that can have been none other than the infirmary. It is at present used as a cow-house! There has been an entrance beside it. The piece of wall shewn at м on the plan is modern, and built of old worked stones. Of the other buildings there are several rooms adjoining the abbot's lodging, and on the north side of the smaller cloister (I, K, and L); these were for offices, and are of early date.3

The tower at the present entrance has been always called a gate-tower; it is necessary, however, first to prove that the entrance was there. Of this I am not at all certain. There are no evidences of any gateway adjoining it, and it may just as likely have been a dove-cot.

Some of the old walls of the grounds to the south are curiously loopholed at Q in the plan. The dotted lines at P indicate the position of former walls. These are likely to have been the various workshops.

At H at the end of the wall in the rear of the abbot's lodging, is an ancient garderobe.

1 Monasticon, v, p. vi of introduction.

2 Dictionnaire Raisonnée de l'Architect. Française, i, 258.

3 Since the Congress, some of the facing has been removed, and very early doorways have been discovered.