THE INTRICATE NETWORK BETWEEN KEPEK-UJ AND AJAGH-ARGHAN. 189
prising to find cataracts at the points where this long chain of lakes restore their water to the Tarim. All the lakes lie at the same level, but the river has a gentle fall; consequently the fall in the case of the lakes becomes concentrated at the spot where their outlet seeks the river. And yet this does not apply to the entire series of lakes, because, as we have seen, the various individual sheets of water are connected together by short waterways, in which there are lively currents. In the lakes themselves, too, there is a movement of the water, although it is very gentle; so gentle, in fact, that their surfaces may be taken to be horizontal.
Thus we were once more out on the broad river. It still presented the same appearance that it had done before we entered the lakes-narrow, deep, extremely sluggish, the water very limpid, after having filtered through the reeds and rushes of the lakes. The alluvial deposits were very small and thin. At intervals a solitary poplar. On our left we passed a hut, long ago deserted, and known as Alim Niasi Iliasidake Muhamede-uji. On the same side too is the not inconsiderable marginal lagoon of Putalik-köl; and on the right the district of Muhamede-mutschekan-uji.
Fig. 155. PASSAGE BETWEEN TWO OF THE LAKES.
June Ist. At our camp we obtained the subjoined measurements: — breadth, 33.7 m.; mean depth, 1.114 m.; mean velocity, 0.5728 m. in the second; and volume, 21.5o cub.m. in the second. The last preceding measurement, plus the Adoke-kokalasi, gave a volume of 20.4 cub.m.
During the day the river described several tiresome little bends, and had a sluggish flow. The barren desert only approached it at a single point. The Putalikköl, on the left, soon came to an end, issuing by a broad and imposing kok-ala, or channel, into the river. But the water in the channel was clear and stationary,
IIIF Color HighRes
