Bodhi- fails to show more than the barest outlines, and these, too, only above the narrow waist. The
sattva-like head had practically lost all modelling and shading over the pink oval which marks its position. figure.
Traces of a nimbus remained. Below the neck there was shown a broad carcanet-like band, in red enriched with jewels and of a type common in both pictorial and sculptural representations of Bodhisattvas since the Gandhàra period of Buddhist art. A closely fitting under-robe of light yellow covered breast and right arm. A dark red garment was visible from above the waist down to about the level of the knees. From the left shoulder there descended a cloak-like upper garment of brownish purple. Lower down, about the left knee, yellow folds were visible, probably belonging to the under-robe. Everything below had disappeared owing to the plaster having broken off. Over the right shoulder traces appeared of another head with an oval band below the neck. The colour scheme in the dress of the figure above described and in what remained on the wall at the back of the passage i distinctly differed from the somewhat crude colouring in the two friezes on wall P.
Remains of The wall at the back of the passage i, over 4 feet thick, retained small portions of its original
paintings painted plaster surface both above and on either side of the arched doorway, 8, seen on the left
behind p p y'
passage in Fig. 467. As this surface is flush with that which bears the fresco last described on wall ß, it
Gha. i. is certain that the mural decoration on both sides of the corner is coeval. From the fact that the
painted plaster survives only from about a foot above the arch it may safely be concluded that the doorway was opened later. As this doorway, 4' 3" wide, does not lie in the centre line of the later vaulted passage, it is probable that it was opened some time before this passage was built. The painted surface to the right of the doorway had suffered too badly from moisture and the mud deposit of white ants for any connected design to be recognizable there. Above the arch I could trace only indications of richly decorated dresses in two places and of a rayed nimbus in red. Farther to the left on the same back wall, the SE. wall of the later vaulted passage built against it had offered some protection. When this later wall was removed remains of painted plaster came to light over a width of about 3 feet. The workings of white ants had defaced much of the surface. But to the right there survived portions of a figure about life-size, dressed in a purplish robe, apparently offering a bowl to some figure on the left. Below the badly broken head appeared a broad jewelled necklace. Of the figure to the left only the folded edge of a similarly coloured robe could be made out.
Structural It was of some interest to note that the ends of both the older walls ß and y, where they are
additions now covered by the adjoining front wall of the large battlemented gateway, were faced with white and
changes. plaster. This clearly shows that the apartment of which these walls formed the NW. side had been
in existence before this gateway was built, and for a time long enough to necessitate the repairs which the addition of the outer wall ß indicates. How far this original apartment extended on the SW. beyond the arched doorway subsequently cut through its wall there, it is not possible to ascertain. But it deserves notice that this wall continues for about 24 feet beyond the doorway.
Indications Judging from the character of the paintings on both the walls ß and y it appears to me probable
of Buddhist that the cella, hall, or passage which stood originally in the position partly occupied by the later worship.
structure i was connected with some place of Buddhist worship. That these mural paintings are of an earlier date than the Muhammadan conquest about the middle of the seventh century may also be assumed on general grounds, though with some reservation in the case of the ` Rustam fresco ', which in view of the position occupied by its deified hero in national epic tradition might well have been tolerated even by zealous converts to Islam as of a secular nature. That Buddhism had during Sasanian times and probably before them possessed a footing in Eastern I ran is not subject to any doubt. How far westwards exactly its influence extended is a question which only future