576 OROGRAPIIY OF CENTRAL TIBET.
5280, 4873, 5288, and 4924 m., this last occurring on Rockhill's itinerary; the mean value is 5091 m. Consequently this valley is trenched to the depth of some 35o m. lower than its two containing ranges.
In Tang-la III we likewise have four passes, namely that of De Rhins, 5163 m.; Littledale's pass, which is marked 5899 m. on his map, though Grenard, as I have said, reduces the figure to 5600 m. (and this is nearer the probable true value, although Slider's Hand-atlas accepts the exaggerated altitude) ; then my pass at 5367 m.; and lastly Bonvalot's at 5500 m. (the beautiful round numbers which he generally gives are in themselves suspicious). However the values just quoted yield a mean pass-altitude of 5408 m. for Tang-la III.
For the latitudinal valley between Tang-la II and Tang-la III we have the following altitudes — 4923, 5228, 5237, 526o, and 4819 m., or a mean altitude of 5093 m., almost exactly the same value as for the immediately preceding latitudinal valley. For the latitudinal valley south of Tang-la III we have these altitudes 4698, 5356, 5127, 5063, 507o, 5028, and 5200 m., giving a mean of 5077 m., which again differs but little from the last two latitudinal valleys belonging to the Tang-la _ system.
All the way from the three "fang-la ranges to the southern boundary of the region which we are considering, that is to say to 31° N. lat., the uncertainty which prevails with regard to the course of the ranges is greater than hitherto, notwithstanding that the travellers' routes run closer together. For we have now to take
• into consideration both Bower's loop towards the north and Rockhill's loop towards
the west, besides which my own route, all the way from Camp XLIV, is divided into two branches, of which that to the east is in places double. Unfortunately just along this very stretch Rockhill gives no pass; possibly he did not cross over any of any altitude worth speaking about, just as I myself in the course of my dash towards the south crossed over none but very low and relatively unimportant passes. In this region the altitudes which I measured all range round about 485o m. Bon-valot however gets in the same region a pass at 585o m.; but that is quite in keeping with the general run of his altitudes. Littledale's pass of 5640 m. and Bower's of 5722 m. in this same region are also doubtful. On the other hand Dutreuil de Rhins, still in this same region, traversed a remarkably low district, with altitudes going down to 4450 m. and with a maximum pass-altitude of 5134 m.
From my two itineraries in this part of the country I obtained the distinct impression that the orographical architecture of this region is far less regular than that of any portion of Tibet to the north of it. It is very seldom that you can speak of definitive mountain-ranges running parallel to one another, and equally seldom of lengthy latitudinal valleys. Both ranges and valleys are of smaller size than hitherto, and often the traveller finds himself wandering in a labyrinth of low hills, amid which the eye is unable to trace out any distinct main range. In a word, the conformation is less regular and the ranges frequently run in directions that deviate from the usual east and west. Between the Tang-la system on the north and the vast mountain-system that accompanied us all the way to Ladak on the south, the orographical structure of the "Tibetan highlands is in one word less regular than that of the country which borders that zone on north and south, though that zone itself
IIIF Color HighRes
