CH. XXI DYKE ACROSS DOMOKO YAR 253
cance for the history of the oasis, past and present. Seventeen years before my visit, I was told, an exceptionally big summer flood had come down from the mountains in the bed of the Domoko stream and had converted a small channel, generally dry, into the broad and deep-cut ravine now extending towards Malak-alagan.
There was serious risk of the whole water-supply of Domoko being drawn off into this deeper ` Yar,' where the difference of level would have made it useless for the irrigation of the extant oasis. So after a year's interval, during which the tendency of the stream to be absorbed in the new bed had seriously interfered with the service of the canals watering the fields of Domoko, the local Begs, acting under the orders of the Keriya Amban, whom the prospective abandonment of the oasis threatened with loss of revenue, set about by a joint effort to erect the present dam. About fifteen hundred labourers, drawn from all the neighbouring oases, were said to have been kept at work on it for close on two months. Considering how widely scattered and scanty was the population, and how divergent the interests of the several oases, the collection of so much labour, all corvée, of course, must have been a serious undertaking. Nor could the dam have been maintained in effective condition without large contingents of men being employed annually on its maintenance during the summer floods.
By these efforts Domoko was assured its former supply of ` Kara-su ' (` black water '), or water from springs, which is indispensable for irrigation during the months preceding the summer floods. But in addition a fresh and constant supply was obtained from the springs which came to light near the head of the newly formed ` Yar.'• This led to the formation of the Malak-alagan colony. The steady growth of the latter was attributed to the water of these springs having remained uniformly ample, and rendered the new settlement less dependent than Domoko itself upon the amount of the Ak-su,' or summer floods. It was interesting to note the uniform assertion that the volume of ` Kara-su ' water available for the canals of Domoko had not been reduced by the formation of the new springs. The evident
IIIF Color HighRes
