above the Hindukush passes.2 This resemblance seems to bear striking proof to the tenacity with which ancient folk-lore clings to the mountains, in Central Asia too as elsewhere, notwithstanding all racial changes.3
I became still more interested in this range, so unlike any I had seen on my Central-Asian travels, when on arrival at Mangush Beg's felt tents late that evening information reached me, elicited with some difficulty through my keen-witted camel-man Hasan Akhûn, about the existence of a stone image to be found high up on the southern side of the range. Mangush Bag somewhat reluctantly, as it seemed, admitted a knowledge of it and agreed to guide me to it. On the following clay a short but fairly difficult march took us across the range by the pass called Saghiz-kan-art (Map No. 14. D. 2), at a height of about 8,600 feet above sea-level, and through the gloomy rock-bound canon which forms the approach to it from the north. Some distance to the south of the pass a natural rock-cistern, known by the name of Shait-kâk, made it possible to fix camp. Thence I proceeded on May 13 under Mangush Beg's guidance in search of the reported image. It proved a very interesting excursion, though its length and most of the ground were such that none but Kirghiz ponies could have covered it in one clay.
After climbing a succession of steep spurs along the south-east face of the range we found a small Kirghiz encampment below the highest of them, the Sar-bal (about Io,000 feet above sea-level). Here, too, as everywhere in this range, the scarcity of water was striking. The Kirghiz were dependent solely on what scanty snow, fell during the winter and remained on the ground for three or four months, to a depth of never more than a span. For the rest of the year water could be obtained only from ' Kâks', i. e. rock cisterns, and a few springs which issue below the small snow-beds to be found high up in sheltered ravines between those frowning peaks. An intelligent old Kirghiz whom I was able to examine at this camp asserted that in his youth rain and snow-fall had been somewhat ampler, and these poor alps used then to be visited by more Kirghiz and supported larger flocks than could subsist now on their stunted vegetation.
From below Sar-bal we rode along gradually widening uplands, skirting the foot of spurs which descend from the Kâka-jâde peaks sighted two days before (Fig. '336). Seen from this side, too, they presented a very fantastic appearance. There was a strange fascination in those towering rock pinnacles, and as many of them were manifestly unclimbable, it was easy to realize why old legends placed among them enchanted strongholds full of treasures. I much regretted that want of time did not permit of closer approach and of an attempt to ascertain the geological reasons for these peculiar rock formations. I may add that the exposed rock in the gorges on both sides of the range was mainly a reddish sandstone.
After a ride of over Io miles from below Sarbal we reached the grazing-ground of Chalkoide, at an elevation of circ. 6,7oo feet, and there to my surprise I found a rustic ' Ziarat' occupying the top of a small rocky hillock. High above it on the crest-line of the range, the peaks of which overlook here a pass leading to the Idak-jilga (Map No. 14. E. 2), there stood out a particularly bold crag ; in it pious Kirghiz eyes recognize the image of an ancient hero called Kaz-atâ. But what arrested my attention much more was the discovery of a real stone image in the centre of the rough enclosure of unhewn stones, about 8 feet across, which marked the holy spot (Fig. 341). It proved to be a stala-shaped slab, 2 feet ro inches high and 122 inches wide, rudely carved in flat relievo, and represented a male figure holding a curved sword with the hands crossed over the breast. The head was disproportionately large ; the feet were not recognizable. The eyes and
2 Cf. Legge, Travels of Fd-hien, p. 24 ; Julien, Bllmoires, tradition from early times has attached to the Nâgas dwelling
ii. p. 206 ; Chavannes, Voyage de Song Yun, p. 28. in mountain springs and lakes, cf. Râjalaraiegini, trans].
' For exactly corresponding legends which Kashmir Stein, i. 263-5, note.