The HvrosTYLE Hall of Xerxes. 313 high relief as in the other edifices. Curiously enough, in the ruins of another palace on the esplanade a great fragment of a figure of this description has been discovered (Fig. 152).^ The dimensions, it must be owned, are certainly much below what would have been requisite for images placed in front of the pavilion, in order to bring them in harmony with the proportions of pillars and capitals around. At Hamadan (Ecbatana) there is a lion which in its terriUy mutilated state is still four metres long. These two specimens are enough to prove that work in the round was not beyond the capacity of the Persian sculptor. So far we have accounted for the reasons which have guided us; it remains to add a few explanatory remarks to enable the reader to understand the plates where die whole of our restora- tion is figured. To take them in their order of succession, Plate IV. shows the geometrical elevation of the palace, from which the front porch has been left out, because its pillars, being in the same line as those of the main colonnade, would have covered and ^JSi^ZlSST^^StJ^'^y. concealed them. We have also refrained from restoring the bulls to which reference was made above, for the simple reason that no data exist as to their shape and character. As to the parapet whereon stood these decorative figures, it is hidden by the basement of the parapet. It will be noticed that our arrangement of the pillars in the central hall is one of four, leaving one pillar out of every five. We have placed this residuum in a single row around the hall, thus bringing all its faces in harmony with the minor colonnades at the sides and front (see Fig. 153). The ex- tremities of the main facade are crowned with a bull capital, the animal being represented full face and not in profile, so as to obtain part of the relief out of the entablature and strengthen the angles of the building. In this fashion we get very nearly the aspect which antcc would have, albeit procured by different means. The idea of fortifying the corners by stretching the device of the capital on to the entablature was sometimes resorted to by Greek architects
- Flandin and Costk, Ptru muiemu^ PUtes CXXX., CXXXV.
Digitized by Google
