s may consist in a delicacy, fragility, and tenderness of
material, which must render it utterly unfit for hard work; and
therefore that we shall admire it the more, because we perceive that if
we were to put much weight upon it, it would be crushed. But, at all
events, it is very clear that the primal object in the placing of such
shafts must be the display of their beauty to the best advantage, and
that therefore all imbedding of them in walls, or crowding of them into
groups, in any position in which either their real size or any portion
of their surface would be concealed, is either inadmissible altogether,
or objectionable in proportion to their value; that no symmetrical or
scientific arrangements of pillars are therefore ever to be expected in
buildings of this kind, and that all such are even to be looked upon as
positive errors and misapplications of materials: but that, on the
contrary, we must be constantly prepared to see, and to see with
admiration, shafts of great size and importance set in places where
their real service is little more than nominal, and where the chief end
of their existence is to catch the sunshine upon their polished sides,
and lead the eye into delighted wandering among the mazes of their azure
veins.

Sec. XXXV. LAW V. _The shafts may be of variable size._ Since the value of
each shaft depends upon its bulk, and diminishes with the diminution of
its mass, in a greater ratio than the size itself diminishes, as in the
case of all other jewellery, it is evident that we must not in general
expect perfect symmetry and equality among the series of shafts, any
more than definiteness of application; but that, on the contrary, an
accurately observed symmetry ought to give us a kind of pain, as proving
that considerable and useless loss has been sustained by some of the
shafts, in being cut down to match with the rest. It is true that
symmetry is generally sought for in works of smaller jewellery; but,
even there, not a perfect symmetry, and obtained under circumstances
quite different from those which affect the placing of shafts in
architecture. First: the symmetry is usually imperfect. The stones that
seem to match each other in a ring or necklace, appear to do so only
because they are so small that their differences are not easily measured
by the eye; but there is almost always such difference between them as
would be strikingly apparent if it existed in the same proportion
between two shaf