$m MEMOIR RELATIVE TO 



roractrical obfervations only^ and with extreme accuracy, if the flations 

 he not very remote from each other, and are fo chofen, that their 

 relative difference of elevation fhall be confidcrable. 



tASE' 4th. 



Tm di/lance and hcighth of a known peak, together with its obferv* 



rcd angle of elevation, give the abfolute heighth of the flationpf obferva* 



i tion— -or, if this be known, the prevailing degree of refradion may be 



obtained : which latter it rriay fonaerimes be important to know ; far 



to the weflward for inftance, where the furfacc of .the country und«i*» 



.lates conliderably, or within the mountains. 



■^€ASE,5r&e 



'jLs, by Xome ©f.the foregoing, Jthc true didancc, and relative pofition 

 ^f two or more flations on the plains oilndia, may be corrc<31y fouad, 

 ib follows, that the true pofitions of fnowy peaks, -not at prefcnt knowHj 

 as well as their altitude, may be found, and that fuch peaks will again 

 enable an obfcrver to determine the pofition of any number of Nations 

 on the pUin, or within the mountains. From whcn.ce they may b^ 

 vifiWe, 



It appears, therefore, that the pofitions of fnowy pcak«, already oh'^ 

 Gained by my furvey, are amply fufficient to corrcd the geography g£ 

 a vafl belt of country : the breadth of which, in a foutherly dircifiion 

 fmm the HimdUya, range, averages from one hundred to one hundrejdl 

 Mid thirty miies, and in, length fomewhat exceeds that of the range 

 , i tie If. 



Ths general direclion of the fnowy chain is from W. N. -V/. to-E* S» 

 E. nearly, to which of courfc the belt is parallel, and if from fuch a 

 line even perambulator routes were furveyed in a foutherly dircdlioo, 

 fo 3=s to make but fma 11 angles with the meridian, the error in mea- 



