1821.] Phtenomena of Heat, Gases, Gravitation, 3fC, 403 



however, as I had read the ingenious treatise of Prof. LesUe on 

 Heat and Moisture, and had considered anew the manner ia 

 which SirH. Davy had conducted his experiments, which were 

 about that time published, I saw that the theory and phsenomena 

 still agreed ; and that the circumstances of my investigation pre- 

 cisely coincided with Mr. Leshe's cooling by gaseous pulsation ; 

 while the error that I had committed, with respect to Sir H. 

 Davy's experiments, rested wholly on my not having taken into 

 account the earth's attraction. 



Their temperatures and elasticities being the same, the ratio 

 of the numeratoms of two homogeneous gases appears by our 

 theory to be equal to the subduplicate ratio of the weights or 

 specific gravities of equal volumes.^' Supposing, therefore, that 

 oxygen and hydrogen are homogeneous (the truth of which in 

 oxygen I much doubt), the numeratom of the former will be 

 quadruple that of the latter. So that if two in volume of hydro- 

 gen unite with one in volume of oxygen to form water, the 

 atoms of oxygen will be double in number those of hydrogen ; 

 and the numeratom of the compound gas before being condensed 

 will be the geometrical mean between the numeratoms of the 

 two simples. It has commonly been conceived that two atoms 

 of hydrogen and one of oxygen form a particle of water ; but 

 whether this, or whether the result of our theory, or whether 

 neither of them be true, it is out of our power to determine. 

 Such kind of speculations transcends the corroboration of any 

 experiment yet devised. I simply mention this theorem, which 

 is one among several that I have investigated, relative to the 

 mixture of gases, to give some small idea of the powers of our 

 theory of the universe for unravelling the secret operations of 

 nature. 



By the same theory I have found that if equal portions of the 

 same gas be mixed together at different temperatures F, F,, 

 accounting in degrees of Fahrenheit, according to the indica- 

 tions of the air thermometer, the mean resulting temperature 

 Fn of the mixture, measured on the same scale, and no 

 extraneous force interfering, will be equal to (448 + F) x 

 • / 448 + F, X a 



{ V ^^^ + ^ ) — 448,t supposing F to represent the 



degrees at the lower, and F^ those at the higher temperature. 

 And if the volumes, instead of being equal, be in the ratio of n 

 to 1 when the temperatures and elasticities are equal, then 

 by a mixture at different temperatures, Fn = (448 + F) x 



♦ This theorem was published in a more general form in the Annals of Philosophy^ 

 for July, 1816. 



-|- 448 represents the degree of absolute cold below the zero of Fahrenheit, employing 

 the air thermometer, and estimating in Fahrenheit's scale of degrees. 



2 c2 



