CCXXXVI PRELIMINARY ESSAY.
everywhere very mountainous except near Wakhsh and the country of Akjar which borders on Mank, a dependency of Khutl."1
Further on, in giving a route from Sfighânian (Cheghaniczn) to Wasjird (TI akhshjird) he mentions that the road comes upon the Wakhshab between' nine miles, and thirty miles from the former place, and that the river has here a breadth of three miles.... "From Wakhshjird to the place where the Wakhshab loses itself under a mountain is one short day... .
" On the borders of Wakhsh and Khutl are WakhAn and Saknia, dependencies of the Turks' country. From Wakhân to Tibet is eighteen days. Wakhân possesses very rich silver mines, producing ore of excellent quality. Gold is found in the valleys when the torrents have been in flood .2 ... Musk and slaves are also exported. Saknia is a town in dependence on the Khizilji Turks. It is five days from Wakhân, and its territories border on the possessions of China."3
In spite of the obscurities of these passages we can gather that the feeder of the Oxus which Edrisi's authorities regarded as the main one came from Wakhân, a' country lying in the direction of Tibet, but that it received somewhere before reaching Tarmedh another great branch called the Wakhshab, so great as to be reported in one part of its course to have a channel three miles wide, and which rose in the Turk's country, i.e. at least as far off as the main chain of the Bolor ; also that between those two great branches lay the provinces of Wakhsh and Khutl.
But where do these two streams join ? Wood, the most competent to have settled the question, in his book, as we have seen, takes no notice of the Wakhshab at all. Nor is there any distinct trace of it in Macartney's map, though a tributary of the Oxus which he represents under the name of the 8urkhdb or R. of Karategin, entering the main stream a short distance above its confluence with the Kokcha has by later geographers (e.g. by the author of the map to Russians in Central Asia) been expanded into identity with the great Bolor-Wakhsh branch. But as Wood in his journey from Kila'h-Chap to Jan-Kila'h and Sayad twice passed the mouth of this Surkhab, so good an observer would scarcely have omitted to notice the confluence of a rival Oxus.
The gallant seaman is still more slightingly treated by Kiepert in his map of Asia. That geographer denies entirely the identity of the river
which Wood ascended for thirty miles (as has just been mentioned) from the Kokcha confluence at Kila'h-Chap to Sayad, with that river which
ä Mank is perhaps the Mungkien (or Munkan) of Hiwen Thsang (see Vie de H. T., pp. 269, 422).
2 Wood mentions a torrent in Wakhan called Zerzumen, probably Zar-
Zamin, " Gold-ground." He also says all the tributaries of the Oxus are fertile in gold (p. 382).
s This Saknia does not seem to be ShagnAn of Wood, which is below Wakhân. It appears to correspond to the Shikini of Hiwen Thsang.