THE ABBEY ClU'UCll UF DOKCIIESTKK. 265 it extended thus far, it must have extended very much fnrther. The Norman choir cither stopped ^vherc the Norman strings terminate at e or else reached as far as the present east ends of the choir aisles. The most probable view is that a small choir such as suggested above was originally designed, but that, during the progress of erection, the design was altered, and the choir carried out on a much grander scale, with such little advance of style as the length of time required for carrying out so great a design almost necessarily involved. I ground this belief on two focts, each of which appear to me to prove one half of it. That such an extended choir was carried out at a period not very distant from that of the erection of the nave is shown by the cei'tain traces of it which still remain. But that such a choir was an after- thought, not a part of the original design, is, perhaps, not absolutely proved, but at least rendered extremely probable, by circumstances tending to show that the point {c) where the Norman string terminates, is no arbitrary break, but marks some constructive division of the church. First, it Avill be observed that at this point an entire change takes place in the external wall on the north side. It is not continued of the same width, but the eastern portion is very much thicker, the excess being external. An arch also, having, as Mr. Addington observes, " much of Early English character," is here thrown across the aisle (at/), dividing the original Norman building attached to the choir from the aisle added to the east of it. Again, the course followed by the Decorated architect when the splendid arches of the choir were added, might possibly tend to show that the Norman wall did not continue any further than it does at present. For in that case one does not see why he should not have cut a fourth arch through the part where the round arch has since been cut, rather than leave a blank wall to the great disfigurement of his choir. P'or though the arch across the north aisle would ^ have prevented a perfectly continuous arcade, yet the difficulty might have been obviated by the employment of a more massive pier •' This arcli, as we shall presently sec, at this point refiuired to be clokcd by an is contemporary with the north arca.le, at arch, it would, even if absolutely cimteni- all events part of the same design, though i)orary, have iiad just the same efVcct on perhaps actually erected earlier. Hut if it the design of the arcade as if it had bceu was thought that the difference in the wall found previously existing.
