THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FOURTH SERIES.] 



MA Y 1862. I 



XLVIII. On some Apparatus for determining the Densities of 



Gases and Vapours. By M. V, Regnault*. 



[With a Plate.] 



THE density of a gas or of a vapour at a temperature T and 

 under a pressure H, is the relation between the weight 

 P ; of a volume V of this gas or vapour, and the weight P of an 

 equal volume of atmospheric air at the same temperature T, and 

 under the same pressure H. If the gas or the vapour obeyed 

 exactly the same laws of expansion and elasticity as atmospheric 

 air, within the range of temperatures and pressures in which it 

 retains the aeriform state, the density would be the same at all 

 temperatures and under all pressures; it would constitute a 

 specific character of the substance. 



But experiment shows that this identity of laws does not 

 exist even for the most permanent gases, for those which have 

 not yet been liquefied. It shows further that the divergencies 

 of the laws of expansion and compressibility are greater for 

 vapours, and even that they often continue up to temperatures 

 much higher than that at which the vapour would assume the 

 liquid state under the same pressure. The density of a gas or 

 of a vapour, as just denned, is therefore not represented by a 

 constant number ; it varies with the temperature and pressure, 

 and these variations are often very considerable. 



It is necessary in physics to deline two kinds of densities for 

 gases and vapours : — - 



1. The real densities, which vary with the temperature and 

 pressure; they ought to be determined by numerous experi- 

 ments, in which the temperatures and pressures arc varied 

 within considerable limits. These densities arc represented by 



* Translated from the Annates de Chimieet de Physique, vol. lxiii. p. 45. 

 Phil. Mag. S. 4. Vol. 23. No. 155. May 18G2. 2 A 



