"For 4 or 5 km of their upper reaches, the Topars traverse the flood plain of the Ili, a plain which was formed when the river channel shifted to the right. In the forties of this century, there was already more water in the Dzhideli system than in the Ili channel, and at present the main part of the flow of the Ili River is along this system. Moreover, the stream velocity in the upper part of the Dzhideli branch is now considerably greater than that in the Ili. Consequently, the Dzhideli branch is able to transfer efficiently the sediment entering it, and in addition it is able to develop its channel. At the same time, some of the sediment is being deposited in the regions where the Topar and Ili branches flow into the lake, due to the sharply reduced stream velocity there; thus, conditions for the intake of water into these systems are becoming less favorable" (Khaidarov, 1960, pp.269-270).
It should be noted that the Bakanas branches were formerly the channels of the Ili River. As early as 1902, L. S. Berg pointed out that a number of large and small irrigation ditches (at present dry) branch out from the banks of the Bakanas branches, and that traces of ancient plowed fields are evident everywhere, providing definite evidence that an agricultural settlement existed there at some time.
Unfortunately, however, these traces cannot be dated accurately. Thus, the regular correlation between climatic changes and river migrations established above by us is the only real basis for assuming that the Bakanas channels were "active" during the warm periods (Bronze Age and early Middle Ages).
".. , a study of the banks reveals that the southern shores of Lake Balkhash are gradually but continually being filled in with fluvial deposits, and that at the same time the northern shores of the lake are being inundated. Places where there were inlets along the south shore 30 or 50 years ago have become dry, whereas the northern banks have an appearance typical of places where the land is being washed away and the water level is rising. On a 1903 map the Boichabyl area was shown as an island, whereas now it has become a peninsula. The inlets at the southwestern end of the lake have dried up completely during the last 15 or 20 years. Thus we see that Lake Balkhash is actually a wandering lake, and is very similar in this respect to Lake Lop Nor in Central Asia" (Fedorovich, 1954).
makes a sharp bend toward the southwest and forms Lakes Tengiz and Kurgal'dzhin, which are situated in the desert-steppe region of central Kazakhstan (Figure 5).
During the third cold period (late Middle Ages) the Nura deviated toward the left (equatorward), and its entire flow went through Lakes Tengiz and Kurgal'dzhin; at this time the level of the lakes was high. As Popolzin (1960) has pointed out, even in comparatively recent times Lake Kurgal'dzhin was a vast body of water with a large open-water surface.
During the third (present) warm period, on the other hand, the river tends to deviate to the right (poleward). Consequently, at the northernmost point of the large bend, some of the water of the Nura breaks through and flows, along the shallow depression between the Ishim and the Nura, into the Ishim basin. This has led to a reduced supply of water to Lakes Tengiz and Kurgal'dzhin, and the latter are rapidly drying up.
"The lakes in the Tengiz-Kurgal'dzhin group are water bodies which are in a state of gradual drying up. This drying-up process became especially intensified after some of the water of the Nura broke through to flow into the Ishim River, causing a certain reduction of the level of the lakes" (Popolzin, 1960, p.84).
The Indus. This river has a southwesterly course in its lower reaches and discharges into the Arabian Sea. During the third cold period (late Middle Ages) the Indus deviated to the left (equatorward), and its course was nearly south-meridional. During the present period of clima tic warming the Indus tends to deviate to the right (poleward). This conclusion is corroborated by the findings of Machacek (1961), who decided that the Indus channel is shifting westward during the present epoch.
The Zerayshan. In its lower reaches this river follows a south westerly course. After irrigating the vast Bukhara oasis, the Zerav shan goes dry, coming to within about 20 km of the Amu Darya (Figure 6).
During the first warm period (Bronze Age) the river deviated to the right (poleward). The flow was then along the right branch, the Makhan Darya, which flowed toward the northwest, as far as the Uch Bash hills. At that time a large delta existed there, and a huge strip of land was under water. Primitive settlements grew up along the Makhan Dar'ya and in the region of its delta, the settlers being cattle breeders and farmers.