branches exactly as the thistle
leaf does along its own stem, and with sharp spines proceeding from the
points, as in Fig. XVI. At other times, and for the most part in work in
the thirteenth century, the golden ground takes the form of pure and
severe cusps, sometimes enclosing the leaves, sometimes filling up the
forks of the branches (as in the example fig. 1, Plate I. Vol. III.),
passing imperceptibly from the distinctly vegetable condition (in which
it is just as certainly representative of the thorn, as other parts of
the design are of the bud, leaf, and fruit) into the crests on the
necks, or the membranous sails of the wings, of serpents, dragons, and
other grotesques, as in Fig. XVII., and into rich and vague fantasies of
curvature; among which, however, the pure cusped system of the pointed
arch is continually discernible, not accidentally, but designedly
indicated, and connecting itself with the literally architectural
portions of the design.

[Illustration: Fig. XVI.]

[Illustration: Fig. XVII.]

Sec. XCV. The system, then, of what is called Foliation, whether simple,
as in the cusped arch, or complicated, as in tracery, rose out of this
love of leafage; not that the form of the arch is intended to _imitate_
a leaf, but _to be invested with the same characters of beauty which the
designer had discovered in the leaf_. Observe, there is a wide
difference between these two intentions. The idea that large Gothic
structure, in arches and roofs, was intended to imitate vegetation is,
as above noticed, untenable for an instant in the front of facts. But
the Gothic builder perceived that, in the leaves which he copied for his
minor decorations, there was a peculiar beauty, arising from certain
characters of curvature in outline, and certain methods of subdivision
and of radiation in structure. On a small scale, in his sculptures and
his missal-painting, he copied the leaf or thorn itself; on a large
scale he adopted from it its abstract sources of beauty, and gave the
same kinds of curvatures and the same species of subdivision to the
outline of his arches, so far as was consistent with their strength,
never, in any single instance, suggesting the resemblance to leafage by
_irregularity_ of outline, but keeping the structure perfectly simple,
and, as we have seen, so consistent with the best principles of masonry,
that in the finest Gothic designs of arches, which are always _single_
cusped (the cinquefoiled arch