Page:Archaeological Journal, Volume 5.djvu/277

 PEVERELL S CASTLE IN THE PEAK. 213 trance into the enceinte was at (D), and here was stationed the porter, whose annual ahowance Ave find entered for such a series of years on the Great Roll of the Pipe and the Clause Rolls, The curtain wall on the eastern side, two hundred and eighteen feet in length, seems to be in great measure a modern protection from the danger of the precipice extend- ing all along that side, though doubtlessly built on previous foundations. The first view of the keep shews two of its sides, the north and west, to be stripped of nearly the whole of their facing, which has been used for building or repairing the church of Castleton. The heart of the walls is formed of the lime- stone rock on which they are erected, though the whole of the ashlaring without and within is of a fine grained and care- fully wrought sandstone grit, brought from a distance, and at vast trouble, when the height of the acclivity is considered. Enough however of the facing remains to sIicav that there was a broad and shalloAv buttress at every angle, as well as one in the central faces of the building. The eastern side has been similarly denuded to a consider- able height. A doorway here under a plain double arch in- dicates the former existence of a chamber, unless indeed it was the opening to a temporary flight of wooden stau's to obtain access to the first floor (B). This side is very few feet from a tremendous precipice, and it required great skill and caution to erect any thing so close to the very edge of the rock. A dangerous descent brings us upon the narrow peninsula under the south side of the keep, and owing to the difficulty of reaching it, we may attribute the almost entire preservation of this portion of the structure, the lower part of the central buttress, and some of the talus being all that is destroyed. There is only one opening on this side, and that a thin slit from the second floor. The building seems to have lost nothing of its original height except the coping, if it ever had it, and all the courses run with the most perfect Norman regidarity. At this, one of the most favourable positions, we will endea- vour to ascertain by the united aid of records and the charac- teristic features of the building, the probable period when it was erected. There is nothing in the masonry to favour the idea of its being an Anglo-Saxon structure, as the existence of VOL. V. F f