34 

 III. BAROMETRIC WORK. 



The barometers used were always carefully compared 

 before and after each day's work, and a long comparison, 

 lasting for three days, was made before and after our trip, 

 which gave very good results, and proved the barometers 

 worthy of confidence. There was a constant difference of 

 002 of an inch, which was obtained from the observations 

 made before we started. This was, at times, increased to 

 004 of an inch, but never more. 



We observed the usual elements for barometric measure- 

 ments in the following order : date, time, temperature of the 

 barometer, height of the barometric column, temperature of 

 the air, and the general state of the atmosphere. The meteor- 

 ological record will give the elements used in our corres- 

 ponding observations. We also made it a point to expose 

 our barometers to the free action of the atmosphere, but out 

 of the direct rays of the sun, in the shade either of a mon- 

 ument, a house, or a tree. In all our measurements the lower 

 barometer was brought as close as possible to the base of 

 the mountain, in order to avoid any differences of pressure 

 which might arise from the horizontal distance of the two 

 instruments. For the same reason we preferred, when prac- 

 ticable, a position in a north and south line, in order to 

 diminish the influence of barometric waves passing from 

 west to east. The exact altitude of this lower station was 

 always obtained (when we could not get the height from any 

 reliable source) by a series of observations between it and 

 some place the elevation of which could be relied upon. We 

 were often helped in this by the consultation of the maps and 

 surveys of railroads, to which we were given access in the 

 most obliging way. The tables used in the calculation of 

 the heights from the observations were the Hypsometrical 

 Tables in the " Smithsonian Meteorological and Physical 

 Tables," prepared by Dr. Guyot. 



The influence of the time of the day at which the observa- 

 tions have been taken, an element too often neglected in 

 barometrical measurements, has been corrected by means of 

 a manuscript table by the same author. This table gives the 

 corrections to be applied to the results obtained for each 

 hour of the day, and each month of the year. These correc- 



