SPITI VALLEY, &c. 



239 



breadth, and, instead of an insulated line of peaks, to present consecutive 

 ranges, or detached clusters of summits rising in rivalry, and even trans- 

 cending the hither precipitous cliffs which appear from our plains in such 

 gigantic desolate grandeur. The passes into the interior of the country 

 were observed to run upon a higher line of level, and the streams to ramify 

 many days journey within the snow-girt region, as the branches of the 

 Ganges, Jamna and Cliuudh, or like the Sallej, to pursue their course through 

 the chain, deriving supplies from its northern skirt, and the high land 

 at the back of the peaks, or penetrating more remotely, and receiving 

 accessions from more inward regions and higher table land, as the Indus. 

 In this vast alpine tract, no line has yet been discovered that marks an 

 opposite slope to the rivers, nor have we any grounds of inference for the 

 probable limit of that lofty level, of which the Ganges and Indus, with the 

 Panjdb streams, maybe considered as defining the southern declivity, and 

 the Brahmaputra and Oxus the eastern and western slope ; but nothing is 

 known or conjectured of a northward or north-eastern boundary, and we 

 still remain ignorant of the extent, the altitude, and the nature of the great 

 central platform of Asia.* Lake Mansarovara may be indeed assumed as the 

 highest point of the Indian Peninsula, forming a plane which throws off 

 the great rivers from south-east to north-west, and the base of clusters of 

 peaks insulated between their sources and the northern slope of the 

 plateau, of which all our knowledge is still confined to conclusions 

 from the upper course of the Sallej and Indus, where the basins of 

 those rivers, and consecpu ntly the lowest depression of the soil, have 

 been ascertained to rest upon the zone of fifteen thousand feet, and the 

 table land, thro ut;h which ihcy roll, to rise beyond seventeen thousand. 

 These arc but approximations to the allitudc of the broken plains of 

 Tarlanj, wliich only serve to prolong conjecture as to the extreme verge of 

 the highest lines of level. A\\. the waters from liic norliiward dcllexureof 



* Baron IIumuolut's licscaiclies were not known to tlic uullior, *\licn tliis wixi vuitten. 



