The projection at the south side of the apse was probably one of the responds of an arch against which the vaulting abutted, as in the crypt. Under the east window the curve of the wall has been flattened, probably to afford a better back for the altar, of which the step remains. On the north side is an aumbry, with a recess adjoining it in the side of the buttress; and on the south side is a smaller aumbry, and a piscina with a projecting basin and a semicircular head, the latter cut apparently in one stone. This again is probably one of the earliest piscinae in existence. The curve of the apse is wider in this storey than below, which partly accounts for the fact that the adjoining Decorated buttress protrudes here into the room. There is also a difference in the stone used, and in several other particulars, _e.g._, the two windows here have very little external splay--all of which may or may not indicate a difference in date between the apse in this storey and in the crypt. The hand of Archbishop Roger seems traceable here not only in the external shafts and corbel-table, but also in the trefoiling (externally) of the east window. The two vaulting-corbels at any rate seem to be his, as well as the piscina. The upper part of the apse has lost its semicircular shape and been squared, and some masonry has been thrown across from its wall to the Decorated buttress, the motive having been perhaps to make a better support for the rectangular east end of the Lady-loft. The oak table in this room was probably the Communion-table of the church during the period following the Reformation.