of stairs that had led up to the platform from the east. It is hence very probable that this side of the platform once carried a kind of porch or antechapel, probably built of wood. As regards the general structural disposition of the shrine, the only comparison which suggests itself to me is with the ruined temple M. it at Mirân which I cleared in 1go7.9 There, too, a high rectangular platform of solid brickwork was surmounted by a structure showing a high core of solid masonry. Though too badly damaged for its design to be determined with certainty, this superstructure may well have comprised a Stûpa dome with one or more colossal stucco images set up against its base or drum.
A reconnaissance made on the day following my first visit had shown me ancient graves scattered in groups along the edge of the Sai terrace stretching to the west of the Stûpa-crowned plateau, Ying. i (Pl. 36). The whole of this edge is cut up by small gullies due to erosion by drainage. About half a mile from Ying. i a wide branching flood-bed of the Shindi river has isolated a portion of it altogether and turned it into a separate small plateau rising island-like between two shallow channels. The graves found on the top of this were less likely to have suffered from damp brought down by surface drainage, and there our search was begun on the third day of our stay. On the west the foot of the plateau had been scoured by floods, and in consequence of this undercutting large planks and other pieces of Toghrak wood were found lying below, evidence that graves had been washed away here. The position of some others was indicated by rows of small rotten posts fixed in the ground, from eight to ten to each grave, at intervals of several feet. This arrangement was sufficient in itself to indicate that the origin of these graves was different from that of the graves at L.S. and L.T., explored lower down on the Kuruk-daryà. To the north of the graves here opened by us I noticed a space slightly sunk into the ground and bare of gravel. On clearing its western portion we laid bare first a thick layer of mixed reed and wheat-straw and then below it thick Toghrak beams fixed at right angles to a post. The natural gravel was reached at a depth of 3 feet without any clue having been found as to the character of the structure that probably once stood here. Possibly it may have served, like the half-underground wooden structure at L.H., to house coffins which have since been completely destroyed by erosion.
The grave first examined, Ying. III. r, lay at the northern end of a row, by the side of a small drainage gully. It was found to contain the body of a man, apparently middle-aged. It was covered with a large hollowed-out trunk laid over a coffin of rough planks. The body was laid with the head to the east, an arrangement observed also in the other graves examined at this burial-place. Of the shroud, which appeared to have comprised both plain white silk and a woollen fabric, only very little survived. The head was less decayed than those of the bodies found in other graves, and was removed with its wrapping of woollen canvas for the purpose of expert comparison of its features with the heads from L.S. and L.T. Near the head was found the wooden cup Ying. iii. i. 02-5, broken into several pieces, and a round wooden tray, badly perished, with bones of a sheep. Here, as in the other two graves found adjoining it on the south, excavation proved very troublesome owing to the rock-like consistency of the top layer of gravel and clay, which salt efflorescence had compacted into a kind of cement. About two feet of this layer had to be broken through, slow work in the absence of pickaxes, before softer soil was reached embedding the coffin.
The southernmost grave of the row, Ying. iii. 2, was found to contain a coffin of superior make and quite unusual size, 71 feet long and 3 feet 2 inches across. Stout planks, 2 inches thick, joined by dowels, formed the covering lid, and the sides, equally strong, were braced together by transverse pieces. Within, two bodies were found side by side, much decayed indeed but undisturbed, and with all sepulchral deposits intact. The body on the left, or northern side, was recognizable as that
9 Cf. Serindia, i. pp. 485 sqq., Fig. 120 ; iii. PI. p.
| || |