relief, which
is lost, and thus a doorway indicated; it would then have a couple of
pilasters and a couple of columns on each flank.

In classic architecture we find nothing that can be compared with this
curious notion of placing columns and pilasters on the backs of real or
imaginary animals, on a lion, a winged bull, or a sphinx. In the modern
East, however, it is still done. The throne of the Shah, at Teheran, is
supported by columns which, in their turn, stand on the backs of lions.
Singularly enough the same idea found favour with European architects in
the middle ages, who often made use of it in the porches of their Christian
cathedrals.[275] Hence, the old formula often found in judicial documents,
_sedente inter leones_,--sitting between the lions--which, was used of
episcopal judgments delivered in the church porch. In Italy, in buildings
of the Lombardic style, these lions are to be found in great numbers and in
this same situation. At Modena there is one in the south porch of the
cathedral that strongly reminded me by its style and handling of the
figures now existing in Cappadocia, of the lion at Euiuk, for example; in
both instances it is extended on the ground with its fore paws laid upon
some beast it has caught.[276] We could hardly name a motive more dear to
Oriental art than this. Between the predilections of the modern East and
those of Assyria and Chaldaea there are many such analogies. We shall not
try to explain them; we shall be content with pointing them out as they
present themselves.

Various facts observed by Sir Henry Layard and the late George Smith, show
that the column was often employed to form covered alleys stretching from a
door to the edge of the platform, doubtless to the landings on which the
stepped or inclined approaches to the palace came to an end. Sir Henry
Layard[277] found four bases of limestone (Fig. 82) on the north side of
Sennacherib's palace. They were in couples, one couple close to the palace
wall, the other in a line with it but some eight-and-twenty yards farther
from the building. In each pair the distance from centre to centre was 9
feet 3 inches. With such a width the covered way may very well have been
roofed with wood, a hypothesis which is supported by the discovery, at the
same point, of the remains of crude brick walls. The columns would mark in
all likelihood the two extremities of the passage. As for the other
conjecture thrown out by the explorer, it s