Page:Further India; (IA furtherindia00clif).pdf/237

dinary level, was estimated by Garnier at 9,000 cubic metres per second, whereas in flood time, on September 20th, the volume was increased to 50,000 cubic metres per second, although both the Se-Kong and the Tonli- Repu fall into the Mekong below this point. The esti- mate for the river at Pnom Penh; in seasons when the river is full, is between 60,000 and 70,000 cubic metres, while Garnier's estimate for the Mekong at Bassak at dead low water was only from 2,000 to 3,000. On the other hand the Irawadi, at the head of the delta, is estimated at 2,130 cubic metres per second, while the waters of the Ganges at a similar point, and at high water, is estimated at no less than 167,000 cubic metres. With this the Me- kong can, of course, make no comparison, yet the rise of the river from low to flood level between Kratieh and Pnom Penh is at least forty feet in the course of the year, a fact which accounts for the constant changes wrought in its bed, and for the immense inundations which serve to enrich and fertilise so large an area of its valley.

As regards the navigability of the river, Garnier arrived at the conclusion that it was feasible for shallow-draught steamers as far as the Sombor rapids, which are at a dis- tance of nearly 400 miles from its mouth; that above this point big poling-boats could be used for its ascent, and large bamboo rafts for its descent; and that below Bassak the Khon rapids presented the only really serious ob- stacle to navigation. Here, however, even if a safe chan- nel could be found, the force of the current was such that no steamer could possibly, he thought, make headway against it.