358 • . THE TSCHERTSCHEN DESERT.
other adjacent parts of the crest of the same accumulation. Now this is in contradiction to the law which I have enunciated, that the rate at which the dunes advance is proportional to their mass. It must not however be forgotten, that other factors may possibly impair the validity of the law. Take the projection of a dune-length, seen foreshortened in perspective (fig. 307), the crest at the transverse section abc is considerably higher than the crest at a, b, cI, and consequently is more exposed to the wind. Here then the wind puts forth greater energy, and its effect may perhaps countervail the retardation of advance which otherwise would be entailed by the difference of mass. I do not however venture to speak with any degree of positiveness upon this point; for after all it may have been an optical illusion which led me to suppose that. the overhanging leeward wall of sand was highest at the thresholds. In a compound dune-length with only one store Wage) of dunes (see fig. 302) the parts of the leeward side which are most advanced are c and c, and these are also the lowest.
Farther in towards the middle of the desert there exist many irregularities in the relation which the leeward dune-wall bears to the threshold below it. In three places, for example, between bajirs 3 and 4, we even find a cessation or shallow breach in the wall (see fig. 305). Between bajirs 4 and 5 we have a bifurcation of the advanced part of the dune-wall (fig. 306), giving rise at B to a smaller bajir, though it is for the most part sanded up. In fig. 308 we have a rough sketch of the situation in bajirs 13, 14, and 15, where the large threshold, interrupted by two small level patches, conjoins with the steep dune-wall to form a distinct re-entering angle towards the east; whence we may infer that the formation of similar angles pointing to the west is not the rule at these thresholds. Fig. 309 shows the breach in the dune-wall between bajirs i 5 and 16; as will be seen, the wall that terminates beside the former tails out towards the south, until it merges in and disappears amongst the dunes. The next fig., 310, represents the complicated situation at the southern end of bajir no. 20, where there were depressions on both sides of our route that
IIIF Color HighRes
