group
of minor piers. The distinction, however, with which we are concerned is
not that of slenderness, but of vertical or curved contour; and we may
note generally that while throughout the whole range of Northern work,
the perpendicular shaft appears in continually clearer development,
throughout every group which has inherited the spirit of the Greek, the
shaft retains its curved or tapered form; and the occurrence of the
vertical detached shaft may at all times, in European architecture, be
regarded as one of the most important collateral evidences of Northern
influence.

Sec. IX. It is necessary to limit this observation to European
architecture, because the Egyptian shaft is often untapered, like the
Northern. It appears that the Central Southern, or Greek shaft, was
tapered or curved on aesthetic rather than constructive principles; and
the Egyptian which precedes, and the Northern which follows it, are both
vertical, the one because the best form had not been discovered, the
other because it could not be attained. Both are in a certain degree
barbaric; and both possess in combination and in their ornaments a power
altogether different from that of the Greek shaft, and at least as
impressive if not as admirable.

Sec. X. We have hitherto spoken of shafts as if their number were fixed,
and only their diameter variable according to the weight to be borne.
But this supposition is evidently gratuitous; for the same weight may be
carried either by many and slender, or by few and massy shafts. If the
reader will look back to Fig. IX., he will find the number of shafts
into which the wall was reduced to be dependent altogether upon the
length of the spaces _a_, _b_, _a_, _b_, &c., a length which was
arbitrarily fixed. We are at liberty to make these spaces of what length
we choose, and, in so doing, to increase the number and diminish the
diameter of the shafts, or _vice versa_.

Sec. XI. Supposing the materials are in each case to be of the same kind,
the choice is in great part at the architect's discretion, only there is
a limit on the one hand to the multiplication of the slender shaft, in
the inconvenience of the narrowed interval, and on the other, to the
enlargement of the massy shaft, in the loss of breadth to the
building.[38] That will be commonly the best proportion which is a
natural mean between the two limits; leaning to the side of grace or of
grandeur according to the expressional intention of the wo