CANFORD CHURCH. 147 



in the former instance carved, and in the latter cut in half to 

 be faced by Perpendicular work, and judge how the operations of 

 the older builders were interfered with by the more modern. 



Then, as regards their chancels, the development of doctrine and 

 the elaboration of ritual connected with the Altar services during 

 the middle ages, necessitated more space. Thus, at that noble ruin, 

 Fountains Abbey, and at stately Durham Cathedral, the chancels 

 are of a later date, and enclose a far wider area than the original 

 Norman constructions. Then, lightning and fire have conspired to 

 demolish, at least, in part, as many of our more modern, so a 

 proportion of our Norman churches. Sherborne Abbey furnishes 

 a familiar example of the destructive work of fire, partially con- 

 suming a handsome Norman edifice. So when we find, as here, a 

 building, the greater part of which may be attributed to the 

 Norman period, as far as its walling is concerned, we are apt to 

 prize it as something precious and noteworthy. 



I think there is conclusive evidence that the chancel, and portions 

 of its adjuncts, the tower, and the nave, with its aisles, are 

 Norman, but, I should say, not of the same period. In giving an 

 opinion on this point I feel considerable diffidence, as becomes an 

 amateur, who never has made a serious study of Church architec- 

 ture, and just knows enough about it to be aware how difficult it is 

 to discriminate between original work and later additions, and to 

 pronounce upon the characteristic mouldings of each style. 



Hesitatingly, then, I would pronounce the chancel portion to be 

 of older date than the body of the church. The arches opening 

 out of the Sacrarium are very rude, springing from square piers, 

 and from the plainest possible imposts. And the same remark 

 applies to those connecting the chancel with its aisles. But the 

 arches between nave and aisles, and those which form the doorways, 

 are considerably enriched, not, however, with the ordinary Norman 

 mouldings of the later date the zigzag, the billet, the chevron of 

 which there is no specimen here, but with a deeply chamfered roll- 

 crease moulding, which is turned up at its terminations on the 

 arches, effectively. Now, the richer mouldings denote a later period 



