838 The American Naturalist. [September, 
submerged. The Madredane channel, three miles long, is narrow, 
and so choked with reeds and aquatic vegetatation that a passage has 
to be hewn through it. The Mosella mouth is little known, but issaid 
to have a bad bar at its junction with the mainstream. The Melamba, 
Maria, East Luabo, and another mouth seem to be closed to naviga- 
tion, as the sea always breaks over their bars. The Chinde mouth 
has, however, three fathoms on the bar at low water, and a channel 
soo yards wide, and well defined. In an exceptionally dry season 
20 to 23 feet were found on the bar at high water. There is a sandy 
point at the very mouth, and good anchorage inside it. Chinde vil- 
lage is at the junction with the main stream, ten miles up. The banks 
and channel of the main stream are continually changing direction 
under the influence of the immense body of water, full of vegetable 
matter, and depositing a light-colored ooze. The delta is thinly 
peopled, and the inhabitants are not indigenous, but have been slaves, 
and are of low social condition. 
M. Dauvergne’s Journeyings.—M. Dauvergne’s ‘explorations 
in the vicinity of the Hindu Kush, last year, led to several interesting 
geographical discoveries. He descended the valley of the Lung, and 
asserts that that river is a tributary of the Tashkurgan, and not of the 
Zerafshan. The valley is deep, difficult of access, warm, highly culti- 
vated, and inhabited by Sunnite Mohammedans, who are Chinese sub- 
jects. The river flows west and southwest, with the Kundur mountains 
on the left bank, and the Kichik-tung on the right. Crossing the 
Kotti-Kandar pass (16,350 ft), which has a glacier on the top, our 
traveller descended into the valley of the Tashkurgan, and then 
ascended that of Karachunkur. He afterwards camped with nomad 
Kirghises in various localities in an elevated rolling Pamir, resembling 
the Great Pamir, 
Our traveller finds the sources of the Oxus or Amu-Daria near the 
pass of Wakijt-Kul, at a level of 15,500 feet, and states that they are 
fed by three enormous glaciers. To make certain, he followed the 
river for seventy miles. i 
It has been asserted that Karambar Sar, a small lake on the north 
side of the Hindu Kush, has two outlets, but one of the results of M. 
Dauvergne’s explorations has been to dispel this idea. There are here 
two lakes in close proximity, the one the real Karambar Sar, about © 
mile and a half long, giving origin to the Karambar or Askaman River, 
while the other, situated a few hundred yards to the east, over ? low 
rocky watershed, is about half a mile long, and gives outlet to the 
Ausa or Marghab. The smaller lake is named Gazkul. 
