Saint-worship increased; pilgrimages increased; pilgrims came in thousands and tens of thousands. They could not be accommodated in the crypts as before: room had to be found on the floor of the choir for shrines transferred from the crypt; and aisles had to be constructed round the shrines, that there might be a free passage, and no dangerous block in the stream of pilgrims. For the local Saint—the St. Thomas of Canterbury, the St. Hugh of Lincoln—accommodation on a vast scale had to be provided. But beside the local saints, there were the great saints of the Church; for them special chapels, with altars, had to be provided, either in a new eastern transept or in aisles added to the central transept. There was, moreover, especially in the first half of the thirteenth century, a great outburst of Mariolatry. For these reasons, then—what we may call ritualistic reasons—vast eastern extensions were made in nearly every cathedral. But the original Norman cathedrals were not only small and inconvenient to the east, but they were throughout very badly lighted; a very large amount of history, as I have tried to point out in speaking of Gloucester, Hereford, and Norwich, consists of attempts to improve the lighting of the cathedral. Sometimes, indeed, the improvement took the shape of total destruction of the old gloomy church, and its replacement by a brilliantly illuminated successor, as at York. Connected with this was the mania for an increased acreage of stained glass—an æsthetic motive which, however, had its practical side; the stained glass justifying itself to the monks and canons as providing a series of lessons in Scripture history or Church history. Many changes were due to damage from fire or storm, or to jerry-building. The clerestory of Norwich had to be rebuilt because it had been crushed by the fall of the spire. The Norwich monks were driven to fireproof the whole church, because fires in 1463 and 1509 showed them the necessity of it, by burning down successively the wooden roofs, first of the nave and choir, then of the transept. I am sure that if our documentary evidence were not so deplorably incomplete, we should find that very many other alterations were due to the effect of fire and tempest: in which case the list of æsthetic changes would be yet further cut down—_e.g._, we might find after all that York choir had to be rebuilt for some purely matter-of-fact reason.