Sec. iii] A HILL RANGE IN THE TAKLAMAKAN 89
formations, has been convincingly proved by the researches of Professor J. Walther." The conditions which are at present progressively wearing down that ancient range across the Taklamakân are likely to have been determining factors all through those ages which have seen it in course of being broken up and reduced to its present insignificant dimensions. Variations of climate may have retarded or accelerated the process ; but in its essential features it must always have been the same, ever since the Tarim basin became what it is, a huge drainage-less area cut off from connexion with the ocean.
The explanation here conjecturally proposed helps us to trace a likely continuation of the ancient range to the south-east, much farther away than the Mazar-tagh on the Khotan river. I have noted elsewhere that the Mazâr of Imam Ja`far Sa.diq, the famous pilgrimage place at the termination of the Niya river, rises on a low isolated hill surrounded on all sides by dunes l' This hill, or rather flat-topped ridge, is covered with stony detritus overlying reefs of rock-salt. I regret not to have brought away any specimen of the decomposed rock, and that the geological formation of the ridge remains for the present uncertain. But reference to the general map of Serindia shows that this denuded rock island lies almost exactly in the direct continuation to the south-east of the line passing through the Bel-tagh and Ökur-mazar-tagh, and only a little to the north of the line connecting the Mazar-tagh of Marâl-bâshi with the Mazar-tagh on the Khotan river. May we not have here the final isolated remnant of the same ancient hill-range striking across the Taklamakân, and completely decayed in this last portion ?
The analogy presented by the survival of the two Mazâr-tâghs and of the Imam Ja`far Sadiq hill, in each case above the west bank of a large river course, suggests the further query whether remains of the same ancient range may not be traceable also in a corresponding position west of the Keriya river, say between Chal-öghil and Yoghan-kum (Map No. 13. D. 4). I regret not to have looked out for such traces when moving along this portion of the course of the Keriya river in 190I and 1908. But I should not feel surprised if the exceptionally high sands, appropriately called Yoghan-kum, were found to overlie some decayed hill formation. This would account also for the delta of the dying Keriya river having its head just at Yoghan-kum.18
Hill above termination of Niya river.
LIST OF OBJECTS FOUND IN DESERT SOUTH-EAST OF CHOK-TAGH
C. xxv. 02. Stone point (arrow-head), Neolithic ; very delicately tapered and rather finely worked. Rhombic section flattened at broad end for junction with shaft. Dark grey chert, superficially disintegrated in places. Found on surface of dusty soil, c. 4 miles SE. of C. xxv. Length 2$", width at base b", gr. thickness A". PI. XXII.
C. xxv. 03. Flint (?) point ; curved, retouched on one edge. Dark horn colour, semi-transparent. Broken at base. Same provenance as preceding. Length 1 i s", gr. width -is", gr. thickness s".
C. xxvii (E. of). oi. Chert core, conical, showing long narrow surfaces of fracture. Base i" diam., h. 'ïs".
C. xxvii (E. of). 02-3. Two frs. of worked stone ; brown to grey, showing bulb of percussion. Gr. M. r é".
C. xxvii (E. of). 04. Fr. of worked stone. Roughly cuboid, one side slightly pointed and showing conchoidal touches. Opposite side broken. •" x i-" x I".
18 See his classical work, Das Gesetz der IPüstenbildung, pp. 6, 32 sq., and passim.
C. xxvii (E. of). os. Fr. of chert core, reddish yellow, from which long narrow flakes have been struck. One side unworked. Length â", thickness I".
C. xxvii (E. of). 06. Fr. of stone, hard, grey-brown; granular fracture, as C. xxvii. o19. Roughly worked. Edges very sharp and clean. Irregular triangular shape. 2i X I" X Iq".
C. xxvii (E. of). 07. Fr. of stone ; hard, grey, shaped like lower end of scabbard, perhaps part of hone. Rounded at one end. 3I" x 1" x a" tapering to i s".
C. xxvii (E. of). 08. Fr. of stone ; hard, greyish-brown, granular fracture as ors, &c., showing bulb of percussion, Li" X1W"Y#"
C. xxvii (E. of). og. Shells found on dune 3 miles E. of C. xxvii. Probably Limneidae ; cf. C. xciii. 02-9. Gr. M. i 6".
17 Cf. Ancient Khotan, i. pp. 313 sq.
18 See Desert Cathay, ii. p. 410.