1848.] through Ariana and India. 485 



of compasses on Wood's map, to which must be added one half more 

 for the windings of the stream, making a total length of 255 miles. 

 From Ish-kashn to Kundut the valley of Wakhan, according to Wood, 

 is from "a few hundred yards to a mile in width. " — The average width 

 is therefore somewhat more than half a mile, as accurately stated by 

 Hwang Thsang. This is one more proof that the distances and direc- 

 tions of the Chinese pilgrim are correct. 



But there is another interesting point mentioned by Hwan Thsang 

 connected with this identification of Hu-mi with Wakhan that in my 

 opinion adds the last link to the chain of evidence in favor of the cor- 

 rectness of my identification. Hwang Thsang says that Hu-mi was 

 one of the ancient districts of the Tu-ho-lo, or Tochari. Now one of 

 the five tribes of the Tochari was named Hieu-mi, and their chief 

 town was called Ho-mi. From them I believe that the Oxus received 

 its name of Amu. This was no insignificant clan, but a mighty tribe, 

 whose king, Kadphises Hoemo (OOHMO), judging from the numbers 

 of his coins still existing, must have ruled over Kabul, and the Panjab 

 for a long time. 



The mention of green eyes points to a mountainous country, and 

 not to the low banks of the Caspian. For it is a well known fact that 

 in lofty mountain-valleys the inhabitants generally have blue or grey 

 eyes, often inclining to green, as is likewise the case with the same 

 colored eyes in Europe. 



P. 1197. — "Gu-la-chi may be Uch" This is another instance of 

 the French ch being misread as the hard English ch. U-la-shi is no 

 doubt the Urasa district of the Kashmirian history, the JJrasa regio 

 of Ptolemy, and the Bash of the present day, for the district of Rash 

 lies just to the westward of Mozafarabad, and to the north-east of 

 Kashmir ; agreeing with the direction indicated by Hwan Thsang. 



P. 1199. — " Che-to-thou-lo, — Khoozdar." This is a third instance 

 of the misreading of the French ch, and distances and directions are 

 again mistaken. According to Hwan Thsang to the north-east of 

 Che-lan-tha-lo, or Jdlandhara, and across precipitous mountains 

 at 700 li (about 117 miles) was Khiu-lu-to, the boundary of India 

 on the north. Both distance and direction point to the district of 

 Kulu, which as Hwan Thsang correctly states, is " surrounded by 

 mountains, and close to the snowy mountains." Major Anderson iden- 



3 u 2 



