1842.] to Shipke, in Chinese Tartary. 391 



is, that the mercury in the tubes of thirty-two and twenty-six and a 

 quarter inches stood exactly at the same point, although the vacuum in 

 the short ones was not more than three-quarter of an inch, and on ap- 

 plying a candle to the top, the mercury rose a little, whereas had there 

 been the least air, it must have sunk from the expansion, which would 

 have been clearly perceptible in so small a space. 



The largest theodolite was constructed by Troughton, and is gra- 

 duated, both vertically and horizontally, to twenty seconds; the elevations 

 of most mountains subtending small angles were taken with it, and 

 those above ten degrees, were observed either with the sextant or circle 

 and artificial horizon. 



At every camp we tried the height of the boiling point with two 

 good thermometers, which very seldom indeed gave the altitude of the 

 place 300 feet different from the barometer, and had we arrived at our 

 ground in sufficient time to distil water, I have every reason to think 

 the disagreement would have been less, for wherever we had an oppor- 

 tunity of using snow, the coincidence of the two methods was most 

 satisfactory. 



The height of the colossal Tuzheegung, whose summit is almost 

 22,500 feet above the level of the sea, was determined by angles of 

 elevation between four and twenty-four degrees, taken at eight different 

 stations, varying from 9,000 to 19,000 feet in height, and from two to 

 about thirty miles distant from it, and allowing one-fifteen terrestrial 

 refraction, the extreme difference between any two of the observations 

 does not amount to 250 feet. The Kylas Peaks, besides several others, 

 were calculated from many stations at various distances, and none of them 

 differ above 500 feet from one another. The next highest peak to the 

 Tuzheegung is above 21,000 feet, it was seen from Hutoo fifty- three 

 miles distant under an angle of 1° 47', and its altitude deduced from this 

 comes within 200 feet of what the observation at Rogee gives it, where 

 the distance was eight miles, and the elevation about fifteen degrees. 



The altitudes of our stations were calculated by M. Ramond's method 

 above Soobathoo, where the barometer was observed five or six times a 

 day during most part of our absence, and the height of the column was 

 invariably measured from the surface of the mercury. By the mean of 

 a whole year's barometrical observations, Soobathoo was found to be 

 4,205 feet above the level of the sea. 



