[M2.8backL] [M2.8back] 8 [diagrams] a. {& because of b} in their centres, would be easily supported by the shorter span of lintel in a1 (and similarly, weights of masonry which would break the sloping stones in b. would be readily sustained by b1) and the second, that supposing the pillars in a. built of small stones - or even of a soft single stone, the weight of the separate lintels would have a tendency to cause fissures in the direction f f while the stone introduced in a1 equalizes their pressure over the whole top of the pillar. (x) This arrangement is the best possible. It may indeed be asked why the process should not be continued, as in a4. but in this case the superstructure is evidently so much raised that it would have been better to have made the shafts of the pillars longer at once; better because simpler and requiring less material: on the other hand it may be asked, why not dispense with the upper flat headstone and employ only the expanding stone beneath; as a5 (opposite) but, though this arrangement is admissible in the case of the single lintel or close set arch, a1 d1 it is evidently weak and unsafe when the masonry is smaller and bears on the edge of the headstone as in the cases a2 d2 which would become a6 d6 both of them evidently unsafe forms if the weight of the superstructure should happen
[Version 0.05: May 2008]