lOD M» Ramond's Instructions for the Application of [Aug* 



consequence, that the established ratio between the densities of 

 air and mercury must be altered. We easily conceive that two 

 strata of air of equal weight, taken one at the equator, and the 

 other at the pole, or the one at the level of the sea, and the other 

 several thousand feet above it, will occupy more space in the 

 latter situation than in the former, abstracting from all other 

 causes of variation in the density. The same coefficient then 

 cannot serve equally for these different cases, unless it be 

 accompanied by a correction which increases or diminishes it in 

 proportion to the variation of gravity. This correction naturally 

 divides itself into two portions : for the diminution of gravity in 

 the vertical line, it is founded directly on the general law, and 

 extends equally to the weight of mercury and that of air. For 

 the diminution upon the meridian, we find the measure of the 

 correction in the length of the second's pendulum which requires 

 to be shortened in proportion as it is less solicited by gravity. 

 The two corrections have each a separate term allotted to them 

 in the formula of M. de Laplace ; and his coefficient being 

 determined for lat. 45° at the level of the sea, the correction is 

 plus in going towards the equator, and minus towards the pole ; 

 while in the vertical it remains always plus ; only becoming 

 subtractive when we descend below the level of the sea into the 

 bowels of the earth, a case which never occurs, except in the 

 bottom of deep mines. 



One correction more completes the number of those which in 

 the present state of our knowledge are essential to the accuracy 

 of barometrical measurements ; and this, though here treated of 

 after the others, is nevertheless, in the order of the operations, 

 the first to be effected. 



It is evident that we shall but very imperfectly compare the 

 heights at which two barometers are sustained, if we have not 

 carefully observed the temperature of each. The point at which 

 the mercury stands is determined not merely by the pressure of 

 the atmosphere, but by the density also of the liquid which 

 forms the counterpoise to it. Now heat dilates mercury, and 

 dtffiinishes its density. In that instrument then, of the two 

 which is the warmest, the column of mercury rises in the tube to 

 compensate by an augmentation of volume the portion of weight 

 which it has lost. If we try the experiment of placing two baro- 

 meters, of perfectly similar construction, and which agree per- 

 fectly togetner, one in a hot apartment, the other in the cold air 

 without, but exactly on the same level, we shall see the same 

 atmospheric pressure expressed by veiy different heights of the 

 mercury. If we take them one to the foot, the other to the 

 summit of a mountain, we shall readily see that before we can 

 form a correct estimate of the difference of pressure, we must 

 necessarily take into account the difference of temperature ; the 

 correction which this circumstance requires is of the most easy 

 description. It results from very exact experiments that a 



