122 



DETERMINATION OF ALTITUDES BY BAROMETER. 



which were referred to the level of the camp itself. The following tahles exhibit the original 

 data and the results deduced by each of the three different methods of computation. It will 

 be seen that the altitudes of forty-eight stations were determined, the highest of which was 

 2,S09.5 feet above, and the most distant 36.4 miles from, the depot camp. Assuming the altitudes 

 found by the level to be correct, a little calculation will show that the old method, the new 

 method with observed air temperature, and the new method with mean daily air temperature, 

 give mean errors of 29.3, 28.9, and 9.2 feet ; maximum errors of 142, 119.1, and 31.9 feet ; 

 and minimum errors of 1.8, 1.5, and 0.8 feet, respectively ; and also, that about two-thirds of the 

 altitudes determined by the last named method differ from the true heights less than nine feet. 



COMPARISON OF THE DIFFERENT METHODS OF COMPUTING ALTITUDES FROM BAROMETRIC OBSERVATIONS, 

 BASED UPON DATA OBTAINED IN THE SURVEY OF CANADA DE LAS UYAS, BY LIEUTENANT R. S. WIL- 

 LIAMSON, UNITED STATES TOPOGRAPHICAL ENGINEERS, IN 1853. 



Data. — Observations in Depot camp during survey of Canada de las Uvas, 



