96 Mr. H. Heaihcote StatJicm [Jan. 31, 



curvature and all leave the point of springing (the capital) with the 

 same curve and at the same angle, and thus the design of the vault became 

 theoretically as well as practically logical, and no adjustment or humour- 

 ing was necessary to make it look exactly as it was intended to look. 

 The fan vault thus forms a peculiarly significant illustration of the 

 predominance of logic in architectural design, since it is perhaps of 

 all architectural features that which to an inconsiderate observer might 

 seem most entirely the spontaneous result of fancy and artistic taste 

 playing easily with the materials before it, whereas it is in reality, as 

 we see, the final solution of a struggle with a practical difficulty, 

 carried on through many generations. What is, however, a curious 

 fact in regard to the fan vault, is that just as the harmonious treat- 

 ment of the vaulting ribs in regard to design was secured, their 

 constructive function ceased. The fan vault was really an inverted 

 conoid generated by the revolution of the transverse arch upon its 

 point of springing. The vault had been gradually tending towards 

 this form with the multiplication of the vaulting ribs, as seen in Fig. 

 14, representing a late form of vault previous to the fan vault ; and 

 when this latter form was actually assumed it was found that the ribs 

 were structurally superfluous — that it was more convenient to build 

 the whole as a solid arched conoid than to build ribs and fill in the 

 spaces between them. Thus the ribs in the fan vault became only 

 a " survival," and are really a feature of architectural exju-ession, 

 fulfilling the same kind of function 5is the flutes of a classic column ; 

 that is to say, emphasizing the lines of upward growth of the design, 

 and giving variety and play of light and shade to what would other- 

 wise have been a dead and heavy mass. 



But what of the other problem, the securing of those points of the 

 wall against which the thrust of the vaulted roof is collected ? Both 

 the Romans and the early Gothic builders felt the necessity of added 

 strength in the wall at these points ; but the Romans merely applied 

 columns to the wall at these points (Fig. 16), and the Romanesque 

 builders projected the wall in a flat pilaster. But the pressure of 

 the vault upon the walls is an oblique one, tending to thrust the wall 

 outward ; it must therefore logically be met by a feature which should 

 obviously be intended to resist oblique pressure and to express that 

 resistance. It was reserved for the builders of the complete Gothic 

 period, in their desire to find ample abutment for the security of their 

 increasingly daring and adventurous vaulting, to develop the true 

 {esthetic form for meeting this oblique thrust, the buttress (Fig. 17), 

 with its lines sloping upward to meet the line of the vaulting-thrust, 

 and the flying buttress, carrying this thrust of the centre vault clear 

 over the roof of the aisle and resolving it into the oj^posing mass 

 of the external buttresses which stand like so many giants round 

 the building to hold it up. Thus the expression of the construction 

 in the design became once more complete and logical, as it had not 

 been since the decease of Greek architecture. And it is instructive 

 to sec how completely this necessity for buttresses to hold up the 



