schools of Northern India fills the place of our slate. This
tablet was found blank, but the marks of plentiful scraping plainly show that it had once been used for writing. In the light of subsequent discoveries we must look upon these few tablets, just like the Indian ` Takhta ' itself, as quasi-archaic survivals. But at the time of their discovery I little suspected what much more extensive finds of the same ancient writing material were awaiting me elsewhere.
The elaborate floral decoration on a portion of a lacquered and painted wood bowl which I found in the same corner clearly betokened Chinese work. And very soon after, as if to confirm a conjecture as to the dwellers in this room, the first finds of Chinese documents rewarded my search. One consisted of a stick of tamarisk wood, about 14 inches long and 1 inch wide, partly flattened on two sides, on each of which there appear in vertical lines about a dozen Chinese symbols. The ink of most has badly faded, and no certain interpretation of the few clear characters has as yet been obtained, though it seems probable that the stick had once been used as a kind of tally, making mention of a certain load. More important from a historical point of view is the second document, a sheet of thin water-lined paper, originally folded up into a narrow roll and recovered almost complete.
According to the provisional translation which Mr. Macartney kindly supplied to me at Kashgar, and which has been confirmed in all essentials by Professor Chavannes, of the Collège de France, the main purport of this paper is a petition for the recovery of a donkey which had been let on hire to two individuals, who after a lapse of ten months had failed to come back or to return the animal. It is precisely dated on the sixth day of the second month of the sixteenth year of the Ta-li period, which corresponds to 781 A.D. The locality from which the petition originates is referred to by a name which, owing to certain doubts as to the phonetic value of the two characters composing it, may be variously read as Li-sieh, Lieh-sieh, or Li-tsa. Now this find is of special value, not merely because it supplies an exact date but also because, in
IIIF Color HighRes
