365 
Suppose I K to be a beam originally in a state of rest, 
against the wall K L. When the beam is moved between 
the planes I L, K L, so that the lower end I slides along the 
plane I L, and the upper end K down the plane K L, the 
centre of gravity will describe a portion of a circle, (the 
centre of gravity being in the middle between the two ends of 
the beam) ; and this may be proved mechanically by the use 
of a small paste-board model. 
Now upon reference to the section of Salisbury Cathedral, 
it will be found that the curve of the sofite (or intrados of 
the arch) of the flying buttress, is regulated by the angle at 
which the buttress is placed against the wall E, and that it is 
a portion of a circle ; also that the thrust of the vaulting at N 
O is carried forward by means of the flying buttress G, into 
the wall buttress H. That is, the direction of the centre of 
gravity is the same as the curve of the flying buttress, so that 
neither the wall E nor the buttress H is distressed. From 
the point P the wall buttress is wholly inactive, and the laws 
of gravitation are left to act unrestrainedly in a downward 
course within the body of the buttress. 
We thus see that the extreme tenuity of the flying buttress 
does not militate against the effective performance of the 
