corona," and recalling probably a "facia" or flat narrow board
such as a carpenter of the present day would use in a similar
position, secured in the original structure to the ends of the rafters
and supporting the eaves. Lastly, the crowning part is, in the Greek
Doric, a single convex moulding, not very dissimilar in profile to the
ovolo of the capital, and forming what we commonly call an
eaves-gutter.

At the ends of the building the two upper divisions of the
cornice--namely, the projecting corona and the crowning ovolo--are
made to follow the sloping line of the gable, a second corona being
also carried across horizontally in a manner which can be best
understood by inspecting a diagram of the corner of a Greek Doric
building (Fig. 57); and the triangular space thus formed was termed a
pediment, and was the position in which the finest of the sculpture
with which the building was enriched was placed.

In the Parthenon a continuous band of sculpture ran round the exterior
of the cell, near the top of the wall.

One other feature was employed in Greek temple-architecture. The
_anta_ was a square pillar or pier of masonry attached to the wall,
and corresponded very closely to our pilaster; but its capital always
differed from that of the columns in the neighbourhood of which it was
employed. The antae of the Greek Doric order, as employed in the
Parthenon, have a moulded base, which it will be remembered is not the
case with the column, and their capital has for its principal feature
an under-cut moulding, known as the bird's beak, quite dissimilar from
the ovolo of the capital of the column (Fig. 65). Sometimes the
portico of a temple consisted of the side walls prolonged, and ending
in two antae, with two or more columns standing between them. Such a
portico is said to be in antis.

[Illustration: FIG. 65.--ELEVATION AND SECTION OF THE CAPITAL OF A
GREEK ANTA, WITH COLOURED DECORATION.]

The Parthenon presents examples of the most extraordinary refinements
in order to correct optical illusions. The delicacy and subtlety of
these are extreme, but there can be no manner of doubt that they
existed. The best known correction is the diminution in diameter or
taper, and the _entasis_ or convex curve of the tapered outline of the
shaft of the column. Without the taper, which is perceptible enough in
the order of this building, and much more marked in the order of
earlier buildings, the columns would look to