54 FROM YASIN TO KASHGAR [Chap. II
near Kizil-tam, an old canal could be clearly made out for a distance of about three-quarters of a mile. At Koshun-kör, a little lower down on the left bank, I found direct evidence that cultivation had actually been resumed since my visit in 1906. This had happened also at the mouth of the Nullah known as Kara-jilga, some three miles above Pisling, the southern limit at that time of Sarikol cultivation. There is thus every reason to suppose that if settled conditions continue in Sarikol, such as have prevailed since 1891, when raids from Hunza ceased, increasing population will lead to the gradual reoccupation of all arable ground right up to the Tàgh-dumbash Pamir.
The great change that has already taken place farther down the valley was impressed upon me when I crossed from my camp at Pisling (11,530 feet above sea-level) to the village of Dafdar, just opposite on the right bank.5 My immediate object was to trace from there the line of the ancient canal known as Farydd ariki, ` Faryad's canal ', famous in local lore and the subject of a curious legend ; I had heard of this canal in 1900,6 but neither then nor in 1906 had I sufficient time to examine it. But my attention was first arrested by the great change in Dafdar itself. Where in 1900 I had seen only a cluster of hovels containing eight or ten families mainly occupied in grazing cattle, I found now some forty homesteads as substantial as those seen about Sarhad. They were widely scattered over an area of well-tilled fields and most of them surrounded by flourishing young plantations of poplars and willows. The total of cultivated ground now extended over more than four miles from south to north with a width for the most part of about a mile. Owing to the abundance of fertile soil and the scarcity of labour the fields are sown only in alternate years with the barley, oats, or peas that form the main crops. Yet there was said to be more than enough water to irrigate all the land taken up, a statement borne out by the luxuriant grass that covered all the waste ground commanded by the canal, as well as by the large stacks of fodder grass stored on the house-tops for the winter.
This striking expansion of Dafdàr was explained by my old Sarikol acquaintance Rashid Beg, who had once more been sent to escort me down to Tâsh-kurghan, as a direct result of the partial reopening of that same ancient canal that I had come to look for. Not very long apparently after my passage in 1900 the military ` Amban ' of Tash-kurghan had effected a clearing of the head portion of the old Farydd-ariki. It was said to have taken him three months' labour with fifty Chinese soldiers and about as many Tâsh-kurghan cultivators. He was energetic enough to camp on the spot and thus saw the work finished in one season. The restored canal, I was told, received its water from the outfall of the Zankan-jilga, which, as the map (No. 3. D. 2) shows, descends from a portion of the snowy range to the south-east, where it rises to peaks reaching about 18,000 feet. The stream draining the valley furnishes not only an abundant supply of ak-su, or water from melting snow and ice, but also a permanent discharge of kara-su from springs.' At the point where I examined the restored canal its alignment ran fully twenty feet above the level of the shallow, carelessly cut channel, then dry, that previously irrigated the then scanty fields of Dafdâr. It carried at the time about 2'75 cubic feet of water per second. The width of the inundated belt on either side of the actual channel, however, clearly showed that the original canal bed had been intended to carry a much larger flow of water, but had silted up since and been only partially cleared. This conclusion was completely confirmed by the subsequent inspection of the ancient canal bed where it had not been touched.
For over three miles we followed the canal winding along the foot of a low conglomerate terrace and irrigating the gently sloping ground towards the river bed. The way in which water was
5 See Map No. 3. c. 2. legend.
6 Cf. Ancient Khotan, i. p. 38. Faryiid is the local Sarikoli 7 For the terms ak-su (` white water ') and kara-su
pronunciation of the name Farhâd so familiar in Persian (` black water '), cf. Ancient Khotan, i. p. 94.