514 GENERAL IIYDROGRAPHHY OF THE TARIM SYSTEM.
was easily able to check, because my guide Kirghuj Pavan saw the stream on both occasions. Thus this river-arm had quadrupled in breadth. This appears to furnish a fresh proof of the tendency that the Tarim shows to flit to the east, a tendency which since my last visit to that region has no doubt become much more pronounced. Tschernoff also reports that the Basch-köl had risen to an unprecedented extent since I saw it in December 1899, that is after its inflow canal was opened. Several of the poplars which we then observed growing on the bank were at the time of Tschernoff's visit sticking up through the ice a good bit away from the new shore-line, which then ran close in under the dune-wall.
Imperfect and fragmentary though these reports are, they are nevertheless possessed of great interest. Not only do they prove that the information which I received about the country below the Ullugh-köl and the Begelik-köl was incomplete, they also witness to the fact that the characteristic marginal lakes embedded in the sand extend, and with the same degree of frequency, all the way down to Arghan, though in point of area they appear to be smaller, and in point of shape more irregular, than those above the lakes just named. Further, these marginal lakes lower down the river would appear to possess a more abundant vegetation. Almost all of them still contain water, although the Tarim is beginning to desert its bed immediately along the front of the sand. From this it may be inferred that the river has shifted its bed at a date so recent that the marginal lakes have not yet been able to dry up, nor yet to grow sufficiently salt to interfere appreciably with the formation of ice. The calculations of area and volume which I made in vol. I for the lakes that I visited are plainly not exaggerated, when it is recollected that those lakes are more in number than I assumed them to be. It is however difficult to say how many lakes have to be added to my list of 35 in consequence of Tschernoff's investigations, for in many cases two or more of his names indicate different basins in one and the same lake. The particulars in this chapter, whatever they are worth, may therefore be regarded as a provisional supplement to my own work.
IIIF Color HighRes
