English and French Barometers. 1 1 



must add two corrections — one for the difference of the tempe- 

 rature of standard, and a second for the decrease of gravity. 

 The corrections will be as follows : — 



Barometer at London observed and! oq.qo^ 



reduced to freezing-point J 



Correction for temperature of standard + "009361 

 Correction for decrease of gravity . -f- '006688 



29-921049 equivalent 

 pressure at Paris. 



This is very close to the equivalent pressure which Regnault 

 gives in his Tables, 760 millims. = 29*922 inches*. 



Of course in a lower latitude the accurate correction for gravity 

 will be greater ; but practically all Tables are made to corre- 

 spond with that of London or Paris ; or else the barometer is 

 supposed to be reduced to the mean gravity of lat. 45°= '0026003. 

 I am somewhat astonished to find that in all the Tables given for 

 the reduction of the barometer to the freezing-point, the only 

 elements taken into consideration in computing the dilatation 

 of the mercurial column are the linear expansion of the brass 

 scales, and the cubic expansion of mercury. Surely for real 

 accuracy the superficial expansion of the glass tube (i. e. increase 

 of its capacity) should also be allowed for, as was done by Mr. 

 Stewart in his experiment to determine the melting-point of 

 mercury (see Phil. Trans, vol. cliii. p. 430). 



Let the expansion in volume of mercury (i. e. cubic) be, as 

 determined by Regnault, ='00018153 for each degree Centi- 

 grade above the freezing-point, and the coefficient of linear ex- 

 pansion of the glass tube be ='0000086130, as determined by 

 Dulong and Petit \ then the coefficient of the superficial ex- 

 pansion [i» e. capacity) of the glass tube will be = '00001 7226. 



Take M for the cubic expansion of the mercury, and G for the 



* I have taken the increase of gravity adopted by Guyot in his Tables 

 = 0'00520048, which assumes the ellipticity of the earth ^oo' 



If we take the ellipticity of the earth as given by the British Ordnance 



Survey, ^ql (see Phil. Mag. vol. xxiv. p. 413), we shall get the increase of 



gravity from the equator to the pole ='00527919, and this will give the 

 difference between that of Paris and London 



= •00023642 = log 4'373684 

 29-905 = log 1-475743 



3849427 = '0070/0, the amount to be added 

 for gravity. 



