cision; and farther, that, if possible, the eye should not be
suffered to rest on the points of junction of the stones, which would
give an effect of instability. Both these objects are accomplished by
attracting the eye to two rolls, separated by a deep hollow, in the
member _d_ itself. The bold projections of their mouldings entirely
prevent the attention from being drawn to the joints of the masonry, and
besides form a simple but beautifully connected group of bars of shadow,
which express, in their perfect parallelism, the absolute levelness of
the foundation.

Sec. IV. I need hardly give any perspective drawing of an arrangement
which must be perfectly familiar to the reader, as occurring under nearly
every column of the too numerous classical buildings all over Europe.
But I may name the base of the Bank of England as furnishing a very
simple instance of the group, with a square instead of a rounded hollow,
both forming the base of the wall, and gathering into that of the shafts
as they occur; while the bases of the pillars of the facade of the
British Museum are as good examples as the reader can study on a larger
scale.

[Illustration: Plate X.
PROFILES OF BASES.]

Sec. V. I believe this group of mouldings was first invented by the
Greeks, and it has never been materially improved, as far as its peculiar
purpose is concerned;[78] the classical attempts at its variation being
the ugliest: one, the using a single roll of larger size, as may be seen
in the Duke of York's column, which therefore looks as if it stood on a
large sausage (the Monument has the same base, but more concealed by
pedestal decoration): another, the using two rolls without the
intermediate cavetto,--a condition hardly less awkward, and which may be
studied to advantage in the wall and shaftbases of the Athenaeum
Club-house: and another, the introduction of what are called fillets
between the rolls, as may be seen in the pillars of Hanover Chapel,
Regent Street, which look, in consequence, as if they were standing upon
a pile of pewter collection plates. But the only successful changes have
been mediaeval; and their nature will be at once understood by a glance
at the varieties given on the opposite page. It will be well first to
give the buildings in which they occur, in order.

1. Santa Fosca, Torcello. | 14. Ca' Giustiniani, Venice.
2. North transept, St. Mark's, | 15. Byzantine fragment, Venice.