Page:History of Art in Persia.djvu/359

 Inhabited Palaces. 339 they are all intact (Fij^^. 14). The character of the entablature is determined by the profile of the indentations cut at the summit of such antoj as have preserved their heads, as well as the marks the bases of the front row of pillars have left on the floor (Fig. 13). By settings up the shafts the real nature of the entablature is revealed to us ; for we then perceive that the position of the capitals was a good deal behind the beds (whose profile; is quite distinct), whereon rested the beam-heads of the roof. Grant(d the relative situation of the summits of the walls and of the supports, the roof must have jutted out far beyond the pillars, penthouse fashion; and M, Chipiez's drawinnrs indicate very clearly the manner the timber pieces were put together so as to bring about this result (Figs. 25, 26).' The deep salience observable in the roof facade is peculiar to Oriental architecture. Greek temples, not excepting the most antique, have nothing of the kind. Conjecture, then, has but a small share in our restoration. If the central crenelation of the landing-place differs from that at the side, this was done for the greater convenience of the inmates of the palace, in order that when they happened to be in the porch they should he able to lean comfortably against the parapet. A lion at rest appears at the summit of the monolith ant^. The massive, pilasterdike character of the latter demanded a crowning member ; and reference has already been made to fragments of figures in the round discovered among the ruins of the Pcrsepolitan palaces, which must have played the part we have assigned to them here in the decoration.* If the figures under notice are not ' The plan of Dieulafoy (L'Ar/ anfi</ue, it., Plate XIII.) brings out the peculiarity in question, as well as in that of Coste, which we figured above (Fig. 13); but Coste did not 6nd out that the penthouse would solve the problem* In the resumtion he has engraved of this buildin'^, he has moved forwaxd the oriamns so as to bring them on a line with the heads of the walls. Nevertheless, he refrained from modifying the result reached by tracings and measurements made on the spot, and allowed the discrepancy between the present state of the ruins and his restoration to remain, so that we aie let into tiie secret of his mward cogitation and consdentioas way of going to work (see i%r»r drMJwsvw, Plates CXXL, CXXI. a.). ' /fisf. of Art, torn. v. pp. 710, 711. A curious instance may be mentioned of a juxtaposition which would help to justify the use w e have made of isolated animal figures, by placing thera at the summit of anta^ or pedestals. A pillar of As(^ the fomons Buddhist refonner of India, carries at the top the image in round boss of a lion (E. Soldi, Les arts mkonnus^ p. 3*6), When Asoka erected his edifices some two hundred and fifty years before our era, the Persejxilitan palaces had not long been overturned. The frequent and lasting relations existing between Persia and India throughout the Achxmenid period, are matters of common Digitized by Google