534 Note on the Construction of the Map of the [Nov. 



Girth! valley has been once explored, I believe, by Manson and Irving 

 in 18 — 1 but without any record of results that I am aware of. My 

 accounts of the Hoti valley between Laptel and Niti were very obscure 

 and contradictory, and in this part of the map there may be great errorl 



The central part of Munshari is studded with a multitude of small 

 villages and hamlets, the spring and autumn residence of the Jwari 

 Bhotias, not half of which are shown in the Atlas No. 66. I have 

 endeavored to supply the defect from information, and my map shows 

 the approximate position of nearly all these places, but they are so 

 crowded together that I was forced to omit the names of many of the 

 hamlets. 



In the trans-Himalayan part of my map, I have copied all of the 

 Indian Atlas No. 65, which shows the explorations of Moorcroft and 

 Hearsay in 1812, taken, I believe, from actual rough Survey of Hear- 

 say's, though not so acknowledged on the map, and the positions there 

 assigned to Gartokh and all the principal villages, rivers, &c. in the 

 route of those travellers, remain unaltered up to longitude 81°, saving 

 the direction of a stream here and there, which I had reason for know- 

 ing to be otherwise. East of that longitude, where the Atlas No. 65 

 terminates, is the result of my own explorations now recorded, includ- 

 ing the lakes with the details of Kailas, and Gangri, the eastern and 

 south-eastward sources of the Sutlej, the sources of the Karnali, Mo- 

 monangli, and the valley of Pruang, with its numerous villages. My 

 survey was a very rough one, made with pocket compass (Smalchalder) 

 and a watch : I took bearings of my course here and there, as I observed 

 any particular change of direction, as also of Kailas, Momonangli, &c, 

 from many different points, and I estimated my distances from noted 

 times by supposed rate of progress according to nature of ground : from 

 the road distances thus computed (at very moderate rates) I made 

 liberal deductions for the map protraction, so that my errors are, I 

 trust, always on the side of diminution rather than exaggeration. As 

 even these rough methods of observation were often interrupted by night 

 marches, &c. the survey is, of course, inaccurate in many respects ; but, 

 at the worst, I suppose that the place which I have assigned to Kailas, 

 the furthest extremity of the survey, lies within a circle of 5 miles ra- 

 dius, described about the true position, and other parts accordingly. 

 Kailas and Momonangli were placed from the average of a number of 



