simply an ornament.

[Illustration: FIGS. 87, 88.--Bases of columns; from the bas-reliefs.]

* * * * *

We have now studied the Assyrian column as a whole and in detail. Most of
its features seem to us to be survivals from the methods and processes of
what we have called the architecture of the tent. The stone column had no
place in those structures of crude brick of which the real national
architecture of Mesopotamia consisted; it was not at home there; the
surrounding conditions were unfavourable to its development. And yet, in
time, it did, as we have seen, put in a rare appearance, at least in the
case of that one of the two sister nations by which a sufficient supply of
stone could be obtained, but even then it filled an ornamental and
auxiliary rather than a vital function. Its remains are only to be found by
patient search, and even in the bas-reliefs its representations are few and
far between. By making diligent use of these two channels of information
archaeology has succeeded in demonstrating the existence of the Assyrian
column and describing its forms, but at the same time it has been compelled
to recognize how narrow was its use, especially in the great structures on
which Mesopotamian builders lavished all the resources of their art. In
those it was employed mainly for the decoration of outbuildings, and it
will be well to inquire how it acquitted itself of such a task.

* * * * *

The column seems to have been introduced in those gateways to which the
Assyrian architect attached so much importance.[272] Read carefully Sir
Henry Layard's description of his discovery of two sphinxes upon one of the
facades of the south-western palace at Kouyundjik (Fig. 83); he gives no
plan of the passage where he found them, but his narrative[273] suggests
the existence of some kind of porch in front of the large opening. It must
have been upheld by a pair of columns on the backs of the two sphinxes, and
may have consisted of one of those wooden canopies which are so common in
the modern architecture of the East.[274]

We are inclined to recognize a pent house of this kind, but of more
complicated construction in the Kouyundjik bas-relief figured above (Fig.
83). No door is shown, but that, perhaps, is due to the sculptor's
inability to suggest a void, or the two central perpendicular lines may
have been joined by a horizontal one on the upper part of the