93 History of Art ik Antiquity. out the fact that we have been tricked, and are the victims of optical delusion. Art in Greece was careful to make the separa- tion of the constituent members of the unit very distinct and visible to the naked eye, so that the spectator should never be puzzled as to the function each was required to fulfil. Here, on the contrary, the base is not infrequently carved into the lower drum of the shaft, and is single with it ; hence with it it must stand or inevitably fall. Elsewhere — ^in the hypostyie hall of Xerxes» for instance — the base is cut in two; in it the torus belongs to the first drum of the shaft, whilst the principal member is a separate block resting directly on the groun d.' Characteristic, too, of this base is a decorative detail that should not go unnotic^. The ornament, unlike that of the Ionian or Corinthian base, where ijt is arranged horizontally, is grouped here in a vertical direction, being in fact but the prolongation and unfolding of the flutes. Despite the elegance of its contour and the care displayed in its make, the base lacks independence, and does not sufficiently con- trast with the column so as to allow of those charming effects which greet us in the Grecian support. The resemblance between the capitals one with the other is greater than that which ciiaractcrizes the bases ; yet here again the Iniildcr did not servilely keep to a un i([ue type, but inodified it here and there. 1 le tried to improve and perfect the primary device he had adopted at first, and strove to introduce some little variety in every proof he drew upon a model whose first imj)rcssion he always kept well in view. The capitals are all zo>ph >ros. The animal that usually appears about the Persian column is a bull,' his legs folded back so as to produce a bold salience at the knee in harmony with the massive head above (I'ig. 32) ; but in the eastern portico of the great Palace of Xerxes, it is replaced by one of those conventional types createtl by Oriental fancy, e.g. a unicorn with lion face, his paws stretched out (1 ig. 31). In ^ FUndin and Coste, Plates LXXXVIII., XCL ; Dieulafov, CArtamtiftu, iL Plate XX. ; Stolze, Plates LIV., LXXV.
- Stoi.ze {Pfrsfpolis, Bfmerkung<n) seems to think that in the capitals of the
columns of porch No. i the animals figured resemble liie horse rather than the bulL Impressions of these fragments are required to verify an observation iriiich no other traveller has made. But we should not be surprised to find that the omamentist hit upon a kind of compromise between the two quadrupeds, so as to add anotlier conventional type to his repertory, which is not a whit more strange than that of the unicorn, found as support to many of the architraves. Digitized by Google
