512 THE ANCIENT BUDDHIST SHRINES OF MIRAN [Chap. XIII
SECTION V.—EXCAVATION OF THE TEMPLE M. v
Character of While the work of packing the fresco panels recovered from the shrine M. III was still in
ruin deter- progress, a closer inspection of the other ruined mounds close by had revealed a small piece of mined.
coloured stucco just to be seen amongst the débris which covered the south face of the square ruin, situated some sixty yards to the north-west (see Fig. III) and marked M. v on the site plan (Plate 29). It was a badly decayed mass of brickwork (Fig. 128), rising to a height of about fifteen feet above the level of the ground, which here showed but slight marks of erosion. The approximately square shape of the ruin and the flatness of its top had from the first suggested that it could not, like some other neighbouring mounds, be the remains of a solid Stûpa. Excavation was started on February 4, and it had not proceeded far before I realized to my great satisfaction that the ruin was that of a temple, in plan exactly corresponding to the one last cleared and containing a Stûpa built within a circular cella.
Structural As seen from the plan in Plate 32, the walls enclosing the cella formed on the outside a square
figures. of over 4o feet. The sides were approximately orientated, with a westward deflexion of about
10 degrees from the true north. The walls were carefully built with sun-dried bricks measuring 16 by 9 inches and about 5 inches thick. They were nowhere less than 7 feet in thickness, this solid construction being, no doubt, due to the necessity of meeting the heavy thrust of the dome which once rose above the rotunda. Outside the cella walls there soon came to light the remains of a square passage which, judging from what survived of the floor on the south (Fig. 128), the least injured side, appears to have been about five feet wide. This floor rested on solid masonry rising about 42 feet from the original level of the ground. Of the wall once enclosing this passage no trace had anywhere survived, and even the floor had almost completely disappeared on the north and west faces. Only on the south had the inner wall of the passage retained enough of its plastered surface to display a small portion of the tempera paintings once decorating it. As seen in Fig. 133, it showed below, in a dado about 1 foot 8 inches high, the boldly painted bust of a winged figure closely resembling the ` angels ' of M. III, and above, in a somewhat narrower frieze, the gladiator-like figure of a man defending himself against a monster of unmistakably classical composition. We shall have to examine the details further on. Here it will suffice to mention that the affinity of the painting in style and design to the frescoes of M. III was so great as to make me certain from the first that this temple dated back to the same early period.
Walls and It was on the south side, too, that the wall enclosing the cella showed less damage than
interior of elsewhere. There it still rose to a height of about Io feet from the level of the floor of the cella, cells.
which was about half a foot higher than that of the outer square passage. In other places it was badly broken, especially on the west side, where a broad cutting, made, no doubt, in the course of treasure-seeking operations of some former period, had destroyed a segment of the cella wall right down to the floor. Even the outer passage had here completely disappeared. On the east, where, as I soon found, the original entrance had lain, the wall flanking it retained a height of only 4 to 5 feet. The north portion of the wall stood to a height of about 8 feet. The Stûpa which occupied the centre of the cella was also broken at a height of a little over Io feet, as seen in Fig. 129 and Plate 32. The circular passage around it, 7 feet wide, was choked by heavy débris of bricks which had fallen from the higher parts of the wall and the vaulting. Its clearing took two days of hard work, though every 'available man was brought up from Abdal to help. But by the evening of the first clay sufficient progress had been made to reveal the dimensions and