104 M, RamoiicPs Instructions for the Application of [Aug. 



with the centigrade scale. It is indifferent by what scale the 

 barometers are divided, so long as the two instruments have 

 similar scales, and the units of the division are again subdivided 

 into decimal parts. The observed readings off of the instruments 

 are to bie first written down. The logarithm of the height of the 

 lower barometer is then to be taken from the ordinary tables. 

 It is at this point of the operation that it is convenient to proceed 

 to the correction for the temperature of the instrument. The 

 table (No. 1) gives the logarithms of these corrections for degrees 

 and tenths of the centigrade thermometer. The difference is 

 positive when the lower barometer is the warmer, which is the 

 most common case ; and negative when the upper. This table 

 gives the logarithm to be added to that of the lower barometer 

 to reduce the insti'uments to the same temperature. We then 

 proceed to take the logarithm of the upper barometer, and the 

 difference of these is the number proportional to the difference 

 of elevation, which we have now to reduce into absolute 

 measures. 



This conversion is effected by multiplying this difference by 

 the coefficient, accompanied by the corrections belonging to the 

 mean temperature of the column of air, the latitude of the place, 

 and the diminution of gravity in the vertical direction. 



We begin by taking the logarithm of the difference of loga- 

 rithms ; to this is then added, 



1 . The logarithm of the constant coefficient for lat. 45° reduced 

 to that sort of measure in which we wish to have the result. By 

 keeping this primitive coefficient separate from the modifications 

 introduced by the latitude, we have a facility of calculating 

 heights in any measures, or of altering the coefficient if it should 

 be found necessaiy. 



2. The correction for the latitude. This need only be calcu- 

 lated for intervals of 1° : a greater degree of exactness would be 

 superfluous. (Table 2.) 



3. The correction for the vertical diminution of gravity, 

 (Table 3.) 



(For the details of these operations, see the remarks and 

 examples accompanying the tables.) 



• In the table of this correction, the first two decimals of the 

 difference of logarithms in the vertical, and intervals of 10° of 

 the thermometer in the horizontal column, are sufficient. The 

 mean differences belonging to each column aftbrd the means of 

 calculating to the third decimal, or to intermediate degrees of 

 the thermometer if we wish. This is generally unnecessary, but 

 these differences have another use, which we shall see as we 

 proceed. 



The correction for the diminution of gravity supposes the 

 lower barometer to be at the level of the sea. So long as the 

 inferior station is much elevated, the correction regulated from 

 this point of departure becomes insufficient to apply to the other 



