are produced by avalanches, and therefore the snow
in them is mixed with rubble, brought down together
with the snow from the surrounding crags.
The valleys of the range are of two principal types,
the transverse with side windings, and the vast longitudinal
valleys, which may be considered one of the
most characteristic features of the Thian Shan.f
It was thought for a long time that the range contained
many volcanoes. The burning mountains of
the Thian Shan were among the “ Traveller’ s T a le s ”
of the last century. Mr. Semenoff, however, declared
that Mount Kullok, south of the Hi,- was not, as
alleged, of volcanic origin, and that the solfator of
Ketu, in the Ili valley, proceeded in all probability
from the combustion of coal.f
Hydrographically the Thian Shan belongs to the
continental drainage of Asia, and in this respect is of
less importance than the neighbouring but inferior Altai
mountains, whence flow the great rivers of Siberia.
Confining our attention for the present to such of its
streams as descend from its northern slopes in Semi-
rechia, I may observe that there is some difference of
opinion as to which are the seven rivers whence comqs
the name of the province. O f those now flowing into
the Balkhash the most important are the Ili, the Karatol,
* The longitudinal valleys generally extend in a direction parallel to
the main range for sometimes hundreds of miles, and besides their
direction these valleys have the peculiarity that their beginnings or
heads are formed of vast river basins, whilst their mouths form rugged
gorges or defiles. In them are nearly everywhere noticeable the most
recent lake deposits, taking the form of horizontal conglomerate and
sandstone.
+ Other explorers, after Semenoff, have denied the existence of
volcanoes in the alleged localities. M. Mushketoff also has come to
the same conclusion, and he supposes that the other pseudo-volcanic
centres of the Thian Shan, that have not yet been examined by Europeans,
may be placed in the same category of solfators.
and Lepsa, all of which I crossed, and to which I shall
allude in the course of the narrative,. whilst there is
another belonging to the province, namely, the Chu,
that ranks next in importance to the Ili.
The principal lakes of Semirechia are the Ala-Kul
— already alluded to— the Balkhash, and the Issik-Kul,
the last being the most interesting as having become
scientifically known to us only during the present
generation. The origin of Lake Issik-Kul is very
problematical. Some suppose that, in a distant geological
epoch, it was part of the general system of the
basins of the Caspian, Aral, and Balkhash. Again,
from the fact that, at the bottom of the lake, buildings
have been discovered, some have supposed that the
locality has subsided, and that the hollow thus formed
has filled with water. Professor Romanoff, however,
in his careful investigations, found no traces of subsi-
dence'of the ground, though the terraces on the shores
show that the waters have receded by evaporation.
Their former level was at least 200 feet higher than
now, and the volume of water is lessening yearly.
Popular traditions say that in the now submerged town
was a well, that one fine day burst forth such a quantity
of water as to engulf both the city and its inhabitants.
The lake is fed by at least 16 streams from the south
and 13 from the north,- but has at present no visible
outlet. A Kirghese tradition says that the natives,
wishing to rid themselves of its waters, dug out the
canal of Kutemaldi, three miles in length, at the western
end of the lake, thereby connecting it with the River
Chu. I f so, however, they made a mistake in their
levels, for instead of letting the waters of the lake into
the river, the reverse took place.
The waters of the lake are now held, as in a cup, in