ON THE PERPENDICULAR OF SOMERSET. 9 



rises two considerable stages above the roof of the cliurch, 

 and is divided into three bays by sllgbt buttresses running 

 up the whole height, and finisbing in pinnacles. The 

 great corner-pinnacles approach somewhat to the Wring- 

 ton and Glastonbury type, but their finish is rather a 

 dorne than a spire ; the domical form comes out still more 

 clearly in the top of the stair-turret. Each bay of both 

 stages contains a long transomed window of two lights. 

 The whole effect is most admirable ; I do not know a 

 more majestic composition of its own class. 



Crewkerne is, in most respects, a grander church than 

 Uminster ; but its tower will not bear comparison. This 

 however partly arises from the arrangements of the church. 

 At Uminster the four arms of the cross are nearly of the 

 same height, the difference being so smaU as scarcely to 

 bear upon the proportions of the tower. At Crewkerne 

 the nave is far higher than the choir, which I cannot but 

 think an inexcusable fault in a cross church, and that 

 one nearly of an uniform date. The result is that the 

 tower from the west looks too low, from the east too 

 elender ; and it has not sufficient merit in other respects 

 to counterbalance this original defect. The part which 

 rises clear above the nave must, I suppose, be considered 

 as forming one lofty stage, as it contains only one long 

 two-light window ; but at the centre of its height there is a 

 set-off in the buttresses, a string along the face of the tower, 

 somewhat like the band in the same position at Mudford, 

 and a break in the window greater than an ordinary transom. 

 The appearance is that of a window which has somehow 

 or other broken through into a stage below its proper 

 one. If I am not intruding on Mr. Riiskin's province, I 

 would compare it to an unwelcome visitor who has thnist 



1853, PART II. B 



