1421), they reached the city of BIKANI on the first day of Rajab (2nd July). Here they were splendidly fêted ; and on the fifth of Shaban (3rd October) they recrossed the Karamuran. Nineteen days later2 they arrived at Kancheu and took up their servants and baggage which had been left there. But they had to halt here two months on account of the disturbed state of the Mongol country ; and they were again detained at Sucheu, so that they did not pass the frontier fortress till some days after the middle of Moharram 825 (about 9th January 1422). Here the whole party were again mustered and registered by the Chinese officials. The troubles in Mongolia induced the ambassadors now to take the unfrequented southern route through the desert. They reached KHOTAN on the 30th May, and KASHGAR on the 5th July. From this they passed the mountains by the defile of ANDIJAN, i.e. by the Terek Daban, and there separated ; one party taking the road to Samarkand, the other " preferring the route of Badakshan" travelled to HISSAR SHADUMAN,3 and thence reached BALKi on the 18th August. Finally on the 1st September 1422 they kissed the feet of his majesty Shah Rukh at Herat, and related their adventures.4
I The dates indicate the position as about one-third of the way from the capital to the passage of the Hoang Ho at Lancheu. This and the name probably point to PINGYANGFU in the province of Shansi, one of the most ancient capitals of China. It is the Pian-fu of Polo, who says
of it moult est grant citez et de grant vaillance ; en laquelle a mar-
chans assez qui vivent d'art et de marchandize. Et si font soie en grant habondance" (Pauthier's Polo, p. 354).
I find that in the identification of the three cities named on the journey through China (Husnabad, Sadinfu, and Bikan) M. Reinaud has anticipated me in every case ; but as my identifications were arrived at independently on the grounds assigned, this is a strong confirmation of their correctness (see his Introduction to Abulfeda, pp. ccclxxxv-vii).
2 Nine days according to the date in Quatremère (14th Shaban), but thus seems much too short. Astley has 24th.
3 The expression in the text seems to show that Badakhshan was sometimes used in a much larger sense than is now attached to it. But this brief indication of the route followed by the ambassadors from Kash-gar to Balkh is particularly interesting, because it precisely retraces Ptolemy's caravan route across Imaus, on the supposition that the Stone Tower was in the vicinity of Ush or Andijan (Andijân=The Stone Tower; Hissar Shaduman=Ascent to Hill Country of the Komedi ; Balkh Bactra). And this is certainly an argument in favour of Ritter's view, for the route from Kashgar via Tashbalik and Wakhsh to Hissar would have been vastly more direct, and there must have been ample reason for not adopting it, even in the height of summer, as on this occasion (see ante, p. cxlix, seq.)
4 I will here insert some remarks on the topography of Rubruquis's travels, in connexion with the site of Equiu .s, which I suppose to be the Asparah of these ambassadors (supra, p. cc).
Rubruquis, riding with Tartars and relays of horses, set out from the Wolga on the 16th September 1253. The route lay straight east, or nearly so, through the country of the Kangli till the 31st October. They then bore a good deal south, passing through certain Alps (mountain pastures ?). On the 7th November they entered a plain irrigated like a garden, through which a large river flowed which entered no sea, but