484 Verification of the Itinerary of Hwdn Thsdng [June, 



tification. As not one of them has stood the test of a rigid examination 

 I consider it clear that the Major's system must be wrong : in further 

 proof of which I will examine a few more of his geographical identifi- 

 cations before proceeding to the historical part of the enquiry. 



P. 1189, " Kiu-mi-tho." — "Kunduz I suspect." Hwan Thsang 

 has just before been detailing the itinerary of the northern bank of the 

 Oxus from Termed eastwards : and beyond Ko-tu-lo or Khutldn, 

 (mentioned above) he placed the mountains of Tsung-ling and Ku- 

 mi-tho, which must therefore be to the eastward of Khutldn near the 

 source of the Oxus ; in which position we find the Komedce Montes of 

 Ptolemy answering to the Tsung-Ling, and the Vallis Komedorum 

 answering to the district of Kiu-mi-tho, Hwan Thsang is therefore 

 right again. 



P. 1189. — " Chi-khi-ni, Cherkes-Circassia," Circassia ! To jus- 

 tify this seven-leagued saltation the Major states that he has no doubt 

 " a leaf has here taken its wrong place." I feel bold enough to ex- 

 press my opinion that the leaf is certainly in its right place, and that 

 Chi-khi-ni is as certainly in the very position indicated by Hwan 

 Thsang. The origin of many of the Major's most erroneous conclu- 

 sions may be noticed in his attempted identifications of this word, in 

 which he evidently reads the French ch as an English hard ch, instead 

 of as the English sh. After correcting this curious " misreading" we 

 have, according to Hwan Thsang, the river Fa-tsu or Oxus to the south- 

 west of Kiu-mi-tho, and the mountains of Tsung Ling ; and to the 

 south of the Oxus, we have Shi-khi-ni or Shakhndn, the Lakinah of 

 Ibn Haukal, and the Sakind of Edrisi : the district on the Shakh-Dara, 

 one of the head waters of the Oxus. 



To the south of Shi-khi-ni, on crossing the Oxus, we come to 

 Tha-mo-si-thiei-ti, or Hu-mi, of which the inhabitants had green 

 eyes. This district Major Anderson identifies with Daghestan on the 

 Caspian : but from the position assigned to it by Hwan Thsang there 

 can be no doubt that it is the present Wdkhdn. The dimensions given 

 to it agree very well with those of the narrow valley of the upper Oxus. 

 Hu-mi was from 1500 to 1600 li (250 to 266 miles) from east to 

 west ; and only 4 or 5 li (rather more than half a mile) in width, from 

 north to south. Now from the Sir-i-kol lake to the junction of the 

 Shakh-dara, the Oxus is 170 miles in length, measured direct by a pair 



