3" History or Artp in Antiquity. esplanade ; each constitutes a porch which is not in touch with the wall, being no more than a passage. But in their character of colossal gateway, four pillars were sufficient support to the roof. In the lateral porches, however, their number is twelve. Except- ing for this, the principle is identical. The great difference observable between the Palace of Xerxes, as we have restored it, and the Hall of a Hundred Columns, though apt to startle at first, is precisely as it should be. No inscription has been preserved of the latter, from which to date this anonymous building, but it certainly was a reception room also, walled on all its faces, with a porch-like colonnade in front (Plates VII., VIII., and X.). As to which of the pair was built last need not be dis- cussed here, but was it likely that so great an effort would have been made for the sole purpose of repeating, on a different scale and with slight variations, an old and familiar theme ? Is it not more natural to suppose that the younger architect wished to create a work that should offer a new aspect? The Hall of a Hundred Columns is but an enlargement of the hypostyle hall, around which chambers are distributed (Fig. 13). The pillared building, how- ever, raised on the verge of the plateau by the order of Xerxes, belongs quite to a different type ; if its dimensions are exceptionally laige, if its size is prodigious and its ornamentation liberal, it is none the less a kiosk. The ground itself shows but one instance which might be taken to favour the hypothesis we traverse. Midway between the front porch and the principal colonnade are four blocks of masonry spaced like the pillars. If good reasons were to hand, for supposing this to have been a walled structure, there is no doubt that these same blocks stand in the situation generally occupied by the main doorways. To those, however, who like us have been led to •form a totally different estimate of the arrangement of the palace there is no difficulty in making out the use to which the foundation stones were put. We may, then, recognize them as the remains of pedestals, separated by wide spaces in gateway fashion. Along with Coste, one is tempted to put colossal bulls on these pedestals, which, agreeably with Oriental tradition, both here, at the Propylaea, and the Hall of a Hundred Columns, acted as sentinels about the doorway.' Nor is this all. As these bulls were not set up against a wall, they ought to have been executed in the round, and not in > Flandus and Cusi e, Perse anden/Uf Plate CXI I.
