484 THE RUINS OF AK-SIPIL AND RAWAK [Chap. XIV
to the bank of the Yurung-käsh is only about seven miles. In fact, to this comparative proximity of the present river-bed were due both the forbidding height of the dunes and the slight depth of subsoil water. The early accumulation of the former had, no doubt, been a chief factor in saving the ruins from destruction, whether from erosion or the hand of man. On the other hand, the presence of subsoil moisture had, as we shall see, adversely affected all materials like timber which were likely to attract and absorb it.
Apart from the question of water, the conditions under which work had to be effected here proved more trying than at any of the other old sites I explored. The season of Burâns had now fully set in ; and the gales that were blowing daily, though from different quarters and of varying degrees of violence, carried along with them a spray of light sand that permeated everything. I noticed the frequency with which the wind would shift round to almost opposite directions on successive days, sometimes even between morning and evening, a feature of Burans well known to all natives living near the Taklamakän 2, and observed also by former travellers. To the discomfort which the constant drifting of sand caused, and which we naturally felt in a most irritating fashion while engaged in excavation, was added the trying sensation of heat and glare all through the daytime. The sun beat down with remarkable intensity through the yellowish dust-haze even while the latter was too thick to permit of satisfactory photographic work, and the reflection of its light from the glittering particles of sand made the heat appear far greater than it really was. The quick radiation that set in as soon as the sun had gone down caused rapid and striking variations in the temperature at different portions of the day. The result of these sudden changes manifested itself in the agues and fevers from which all my followers began to suffer after our start from Yurung-kash. It was impossible for me to escape exposure to these adverse atmospheric influences ; but I succeeded in keeping their effect in check by liberal doses of quinine until my work at these ruins was done.
The excavations which I commenced on the morning of April i i within the south corner of the quadrangle soon revealed evidence that the enclosing wall had been adorned on its inner as well as on its outer face with rows of colossal statues in stucco, thus making the quadrangle correspond to the chapel courts of Buddhist Vihäras in Gandhâra and elsewhere. Those on the inside face of the wall might be expected to be still in a fair state of preservation, owing to the depth of the sand, which even in the most exposed portion of the Stûpa court (between the gate and the south corner) was nowhere less than 5 ft., and greatly increased towards the west and east corners. It was clear that great masses of sand would have to be shifted before these sculptures could be systematically unearthed and examined in safety. For the heavy earthwork implied by this task it was necessary to await the arrival of the reinforcements already summoned. But in the meantime I was able to utilize the dozen labourers already at hand for such clearings as the preliminary survey of the architectural remains demanded.
From the plan on Plate XL which was prepared on the basis of this survey and of subsequent more detailed measurements, it will be seen that the Vihâra court formed a great rectangle, measuring 163 ft. inside on its south-western and north-eastern faces and 141 ft. on its shorter sides towards the north-west and south-east. It was enclosed by a wall about 3 ft. 6 in. thick, solidly built of sun-dried bricks. At the south corner, which of the portions not completely buried under sand was best preserved, this wall rose to a height of II ft. but was probably once still higher. The bricks used in it, as well as in the Stûpa, measured on the
2 Thus, e.g., a strong wind blowing from the north-east on April 13 was followed on the r 4th by one from the north-west, which in the afternoon veered round again to the