Sec. v] THE DATE OF RAWAK REMAINS ; THE JUMBE-KUM SITE 501
was unearthed in a completely rotten condition I. Sticking between the plaster of the base on its four sides and the decayed wood of the boarding, at heights varying from r 2 to 2 ft., were found numerous copper coins, which, as far as they have not suffered too much from oxidization, can all be recognized as wu-chu pieces. On the south side twelve complete coins were recovered, with numerous fragments of others ; on the east side fourteen, with some others which broke into fragments when I attempted to separate pieces which corrosion had stuck together. From the west came twenty-two, with more fragments (for a specimen see Plate LXXXIX, 19), while twelve were discovered on the north face. There could be no doubt that they had been slipped into the places where they were found, through small interstices of the boarding, manifestly as votive offerings, just like the pieces previously discovered in small cavities below images and underneath plaster mouldings.
Most of the coins are in good preservation, apart from the oxidization they have undergone, and do not show any marks of long circulation previous to their deposition. Only current coins are likely to have been used for such humble votive gifts ; and as no coin finds of a later type have come to my knowledge, we are justified in assuming that the latest known date of these coin-issues marks the lowest chronological limit for the abandonment of the shrine. This conclusion is supported by the evidence of the eighteen coins picked up by my men between the dunes near the ruins, ten of which bear the legend wu-chu, while the rest are small square-holed coins without legend. The great find, previously mentioned, of a pot full of wu-chu coins at an eroded spot about one mile to the south-east also agrees with it.
The wu-chu currency belongs properly to the period of the Former and Later Han dynasties. The rule of the second extended from 25 to 220 A.D., but the issue of its coin types appears to have continued in China up to the close of the fourth century, if not to the advent of the Tangs. It is thus difficult to determine the date when the current use of wu-chu coins, whether imported from China or locally coined, is likely to have ceased in Khotan. There remain, however, two significant negative observations to guide us. On the one hand, the fact of the numerous coin finds of the Rawak Vihâra not comprising a single later piece makes it probable that the date of these votive deposits could not have been removed by many centuries from the period of the Later Hans, when the wu-chu coinage was the recognized currency of the Chinese empire. On the other, the complete absence of Sino-Kharosthi pieces seems to preclude the assumption that the shrine had existed in the first centuries of our era. The absence among the votive deposits of the earliest Chinese coin type, that without legend, points also in the same direction. Possibly the complete excavation of the Vihâra may hereafter provide a more definite indication of date. But combining what evidence is at present available in the coins, the style of the sculptures, and such minor antiquarian indications as may be derived from the constructive plan and the materials of the ruins, we can scarcely assert more than that the extant shrine must have been abandoned at some period between the third and seventh centuries of our era.
At the commencement of the excavations I realized with regret that, owing to the extremely friable condition of the stucco and the difficulties of transport, the removal of the larger sculptures was quite impracticable. Those pieces of the colossal images which were found already detached, such as portions of arms, edges of drapery, &c., usually broke when lifted, whatever care was used. An attempt to move the complete statues or torsos from their places without elaborate
A base revetted with timber in exactly the same Niya Site which is likely to have been a shrine; see above,