265 

 TABLES 



SEDUCING THE BAROMETRICAL OBSERVATIONS TAKEN AT ANY TEMPERATURE TO "IIE 

 TEMPERATURE OF THE FREEZING POINT. 



The variations of the mercurial column in a stationary barometer are due to two 

 causes, the changes of atmospheric pressure and the variations of temperature of the 

 mercury, which affect the length of the column by changing its density. The varia- 

 tions of atmospheric pressure, which alone the barometer is destined to ascertain, are 

 therefore hidden, and their observation falsified by the expansion or contraction of the 

 mercury due to changes of temperature. For, supposing that, while the atmospheric 

 pressure remains the same, the temperature of the instrument becomes lower, the 

 mercurial column wiJ become shorter, and the barometer will appear to fall ; if the 

 pressure becomes less, but the temperature increases, the expansion of the mercury 

 will tend to compensate the diminution of pressure, and the barometer may remain 

 stationary, or even may rise, while it ought to be falling ; in other cases the action 

 of temperature will tend to increase the amount of the changes of the barometrical 

 height. It is therefore evident that successive observations, with the same barometer, 

 do not givp directly the actual changes of atmospheric pressure, unless they hav« 

 been taken exactly at the same temperature, a case which, in practice, seldom occurs, 

 liikewise simultaneous obsei-vations, taken with various barometers, do not g"ve 

 directly the actual differences of the absolute pressure of the atmosphere above the 

 instruments. To obtain the true barometrical heights, that is, the action of the at- 

 mospheric pressure alone, the influence of the temperature must first be eliminated 

 from the observed heights. This is done by reducing, by means of the following 

 Tables, the various barometrical columns to the length they would have at a given 

 temperature, which is the same for all. For the sake of convenient comparison, 

 the freezing point has been almost universally adopted as the standard temperature 

 to which all observations are to be reduced. 



Construction of the Tables. 



In all the following Tables the barometers are supposed to be furnished with brass 

 scales, extending from the surface of the mercury in the cistern to the top of the 

 mercurial column. The correction to be applied is therefore composed of two ele- 

 ments : the correction for the expansion of the mercury, and that for the expansion 

 of the scale ; both of which ought to be. and have been, taken into account. 



Indeed, the correction for the expansion of mercury is not sufficient to reduce the 

 readings to the height which the barometer would indicate, under the same pressure, 

 at the temperature of the freezing point. For when the temperature rises the mer 

 curial column expands ; but then the scale also grows longer, and this will tend to 

 Jower the reading of the height. The correction for the expansion of the mercury 



C 61 



