Page:History of Art in Persia.djvu/386

 364 History or Art in Antiquity. difference was the omission from the facade of the two chambers Fergusson has put there to fill triangular hollows exceedingly dis- agreeable of aspect Two years later, we are treated to a totally diflerent conception. True, he has preserved the triple colonnade* but the front portico has been transferred to the rear of the square phalanx. The hall, instead of being entirely enclosed, is walled in on three sides alone, the front portico looking outwards. The latter is comprised within antae shaped upon the models of Assyrian towers.* Which of these two contradictor)- plans should have been selected ? Is that which appeared in 1889 to be final, or are we to look upon it as a mere expedient to show how the hall, which would have been left in utter darkness, could be lighted, or as a means devised for the greater display of the central phalanx and its long ranges of lofty columns ?' The dearth of documents, then, precludes our entering into a discussion as to which of the two restorations is in better accord with the remains of edifices which have been exhumed. This only we would observe. Whatever may be thought of the un- expected scheme Dieulafoy seems to have definitely adopted, it can in no wise influence our restorations of the buildings at Persepolis. In speaking of these it should be remarked that all the plans that have hitherto appeared, invariably place a portico both in front of the hypostyle hall, the throhe-room and the in- habited palaces. No possible doubt exists on this head ; it is as dear as daylight that the principal entrance to both the Hall of a Hundred G)lumns and the Great Hall of Xerxes was from the vestibule, with two ranges of columns flanked by a pair of winged bulls (Fig. 1 59). Whether the latter was open, as the state of the ruins indicates, or walled in as Feigusson assumed, its true entrance was on the north-west face opposite the Propylaea, square pillars. Had ornament of tiiat nature been of fiwquent occurrence in Persian architecture, should not we find traces of it in the Hall of a Hundred Columns, and in the palaces of Darius and Xerxes ? Yet their pilasters are perfectly sniooih ; more than this, part of their mass towards the base — at least in the first- mentioned edifice — is infringed upon by the body of the bulls, thereby excluding the idea of a pand contrived in the foce of the wall ; in that it would have been shorn of its proper height and thus destroyed the effect of the device^ • Dieulafoy could have obtained the same result with less expenditure of time and labour, by presenting a transverse section similar to that which yielded our perspective view of the interior of the Hall oi a Hundred Columns. The wall of the lagade and the portico would have had to come away. Digitized by Gopgle
 * We have our doubts with regard to the channelUngs Dieulafoy has put over his