of the monkish establishment. But thât the population which supported it was not Chinese is plainly indicated by the transcribed names of the borrowers and their sureties, as well as by the short inscriptions in cursive Brahmi found beneath some of the frescoes of the temple. The more learned of the monks may be supposed to have been versed also in Sanskrit, the sacred language of the Buddhist Church throughout the North ; a small fragment of a Sanskrit ` Pothi ' was, in fact, among the manuscript finds of this ruin.
The very pettiness of the affairs recorded in the Chinese papers of this small convent increases to no small extent their value from a chronological point of view. Unimportant in character and insignificant in size and material, it is highly improbable that these documents should date back to a period preceding by any great length of time the final abandonment of the building. Now it deserves to be noted that all the papers from this ruin which can be dated with accuracy, belong to the years 782-787 A.D. Taking into account that the first-described Chinese document, found in the ruined building D.V. under exactly similar conditions, also bears the date of 1.D. 781, we are almost by necessity led to the conclusion that the settlement to which these shrines and dwelling-houses belonged was deserted about the close of the eighth century of our era. In each case the papers were discovered closely adhering to the original floor, which proves that the sand must have entered the rooms very soon after these petty records had been scattered about there. For light and flimsy as they are, the little paper-rolls could not have resisted very long the force of the storms that pass over the country each successive spring and summer.
It is a particularly fortunate circumstance that such unmistakable chronological evidence has been obtained, in the very same structure which has furnished us with the best preserved, and perhaps the most interesting specimens of contemporary painting from this site. The three painted panels of wood I refer to were found lying in the loose sand a few inches above the flooring, and not far from the east
IIIF Color HighRes
