an ox-head, reduced to a small scale and showing two horns wide apart. Its shape closely corresponds to that of Rustam's famous gurz, as Persian iconography of Muhammadan times knows it in familiar illustrations of the Shâhnâmah and elsewhere. The left hand is raised above the waist. What the object is which it may hold cannot be made out.
From elbow to wrist the arm is covered by the same dark blue robe which extends from below the neck to above the knees. This robe is edged with a dark red band showing a pattern of yellow spots like those described above. Only this band is visible on the right arm. A dark red belt, with a jewelled round buckle in the centre, encircles the waist. From below the robe the left leg with bent knee projects, clad in richly ornamented narrow trousers or leggings. Their colour is dark red, on which is shown an elaborate floral pattern in yellow, embroidered or woven. The left foot, turned to the left, wears a dark-coloured boot reaching to the ankle. A narrow sword, with its sheath represented by a red stripe with bead ornament along its length, hangs against the left thigh. [Passing behind the calf of the visible leg it reappears in front of the shin. One of the slings supporting the sword shows just below the waist-belt.] From behind the figure's neck there flutter upwards two curling white bands shown across and beyond the greenish halo which encircles the head. As the head-dress, too, is white, they may, I think, be safely taken to correspond to the scarf-like taenia shown fluttering behind the crown of kings on Sasanian rock sculptures and on coins. Beyond this imposing figure only the head of an attendant could be recognized, drawn on a much smaller scale and turned towards the former.4a
There could be little doubt that the seated figure receiving worship and offerings was meant to represent a quasi-deified personage, and the ox-headed mace in his hand unmistakably pointed to Rustam, the great legendary hero of Sistân. In fact, the men from the hamlets across the ` Naizâr' who were employed by us at once recognized the familiar emblem and spread the news of the discovery widely. But quite as interesting to me, and in some respects more puzzling, appeared the three-headed figure standing in worshipping attitude before the seated one. The treatment of the composite head is exactly the same as in Trimurti representations of Buddhist art known from Chinese Turkestân,5 and this necessarily draws attention to other points of contact with certain Central-Asian Buddhist paintings in details such as those of the jewelled ornaments, coloured haloes, and dress. I could not help being reminded in particular of that fine painted panel, D. vii. 6, brought to light by me in one of the ruined Buddhist shrines of the Dandân-oilik site in the desert N E. of Khotan, and of the strange armed and booted divinity, wholly Persian in style of figure and rich dress, which one side of it shows seated on a flowered cushion. When describing this remarkable picture, deposited in a place of Buddhist worship and yet presenting a figure so curiously unlike those of other Buddhist divinities,' I emphasized the unmistakably Iranian character of this figure, evidently locally adopted into the Buddhist pantheon, but could offer no certain clue for its identification.
Re-examination of this panel in the light of the wall-painting discovered in far-off Sistân seems to me now to supply this clue and at the same time to help us to the right perception of the
4a Mr. Andrews draws my attention to interesting parallels ` The reason for the sword passing behind the lower leg
in the following note : is to prevent it sticking awkwardly out from the thigh when
` The pose of the " Rustam " figure in the lower register sitting—a detail noticed by some observant artist in antiquity
is exactly paralleled by that of the figure of a king (Yezdegird) and adopted thereafter by other artists as a happy touch of
on a silver bowl of the 5th century (Sarre, loc. cit., Pl. iii). realism so dear to the Persians and their forerunners.'
The curious position of the sword is clearly shown, and the 5 See the three painted panels D. vII. 6, x. 5, 8, from
absence from view of the right proper leg is noticeable. In Dandân-oilik, Ancient Khotan, ii. PI. LX, LXI, LXII ; ibid.,
fact both our painting and the repoussé figure would appear i. pp. 298 sqq.
to be inspired by the same original. 6 Cf. Ancient Khotan, i. pp. 279 sq.