§ 1. We have seen in the last chapter how, in converting the wall into the square or cylindrical shaft, we parted at every change of form with some quantity of material. In proportion to the quantity thus surrendered, is the necessity that what we retain should be good of its kind, and well set together, since everything now depends on it.

It is clear also that the best material, and the closest concentration, is that of the natural crystalline rocks; and that, by having reduced our wall into the shape of shafts, we may be enabled to avail ourselves of this better material and to exchange cemented bricks for crystallised blocks of stone. Therefore, the general idea of a perfect shaft is that of a single stone hewn into a form more or less elongated and cylindrical. Under this form, or at least under the ruder one of a long stone set upright, the conception of true shafts appears first to have occurred to the human mind; for the reader must note this carefully, once for all, it does not in the least follow that the order of architectural features which is most reasonable in their arrangement, is most probable in their invention. I have theoretically deduced shafts from walls, but shafts were never so reasoned out in architectural practice.