[3] History of the Cathedral, 125.
[4] The Doulting stone, of which the cathedral is built, comes from the St. Andrew's quarry at the little village of Doulting, where Bishop Ealdhelm died. It is inferior oolite, and very like Bath stone, which is the greater oolite. The exterior shafts were blue lias, and those within either blue lias or Purbeck marble, though there are one or two shafts of red Draycot stone in the western responds of the nave.
[5] Cathedrals, iv. 98.
CHAPTER III
THE INTERIOR
The earlier architecture of Wells Cathedral presents so many puzzles, that the most skilled experts have differed widely both from each other, and, as we know now, from the truth. There are four distinct varieties of Early English work, covering a period of about a century from the time of Bishop Reginald, whose episcopate began in 1174; and yet, until Mr Bennett deciphered the old charters, which have at length settled the problem, all the work was attributed to Jocelin, for nothing was known of Reginald's building, and some of the best judges were even convinced that the west front was built before the nave. The difficulty was mainly caused by the unusual character of the architecture of the nave; "unlike that of any ordinary English building, and belonging to a style on the whole fifty years earlier" than the west front, as Professor Willis said, who gave it a name of its own, and called it the Somerset style. Thus the theory came to be that two bodies of masons had been employed—an ordinary English company for the front, and a local Somerset company for the nave, transepts and choir, who worked in a local variation of the prevalent Early English style. In this way, an attempt was made to overcome the difficulty of attributing to Jocelin work which Mr Willis had himself pronounced to be "only a little removed from the early Norman style." Mr Freeman, too, had allowed that the north porch might be earlier than Jocelin; and, long before, Britton had said that there would be little hesitation in ascribing the church to the transitional period of Henry II. (1154-89) on architectural evidence, were it not for Godwin's assertion, that Jocelin had entirely pulled down the old church and built a fresh one.
But now we have got behind Godwin, and have found from contemporary evidence that Bishop Reginald commenced the present church. Thus we are able to divide the Early English work into no less than four periods, (1) The three western arches of the choir, with the four western bays of its aisles, the transepts, and the four eastern bays of the nave, which are Reginald's work (1174-1191), and so early as to be still in a state of transition from the Norman. It is a unique example of transitional building, and Willis calls it "an improved Norman, worked with considerable lightness and richness, but distinguished from the Early English by greater massiveness and severity." The characteristics of this late twelfth-century work are bold round mouldings, square abaci, capitals, some with traces of the classical volute, others interwoven with fanciful imagery that reminds us of the Norman work of Glastonbury; while in the north porch, which must be the earliest of all, we even find the zig-zag Norman moulding. (2) The rest of the nave, which was finished in Jocelin's time—that is to say, in the first half of the thirteenth century—preserves the main characteristics of the earlier work, though the flowing sculptured foliage becomes more naturalistic, and lacks the quaint intermingling of figure subjects. (3) The west front, which is Jocelin's work, and alone can claim to be of pure Early English style. (4) The chapter-house crypt, which is so late as to be almost Transitional, though, curiously enough, it contains the characteristic Early English dog-tooth moulding which is found nowhere else except in the west window. From this, we reach the Early Decorated of the staircase, the full Decorated of the chapter-house itself, the later Decorated of the Lady Chapel, the transitional Decorated of the presbytery, and the full Perpendicular of the western towers.
Much of the masonry in the transepts, choir, choir aisles, and even in the eastern transepts, bears the peculiar diagonal lines which are the marks of Norman tooling. This does not, of course, prove that any part of Bishop Robert's church is standing, for medieval builders were notoriously economical in using up old masonry, but it does show that there are more remains of his work in the building than was generally supposed. A characteristic feature in this Norman tooling is that if a rule be laid along its lines, they will be found to be very slightly curved, a feature which is due to the fact that Norman masons dressed their stones with the broad curved blade of an axe.