of Air under Volwne and under Pressure. 329 



liave been used as the basis of such calculations, considering 

 that, at as early a period (1812), MM. Desormes and Clement, 

 had, with a very different view, made some better-contrived ex- 

 periments, from which an approach to the true quantity could 

 have been made with more certainty. Their method was very 

 simple, and required no thermometer to shew the variations of 

 temperature, — a contrivance which is said to have been first sug- 

 gested by Lambert. No notice, however, seems to have been 

 taken of these latter experiments, — probably because they were 

 associated with a most fanciful inquiry after the absolute zero, 

 till MM. Gay Lussac and Welter undertook a similar and more 

 extensive series of experiments, giving nearly the same results. 

 Of both of these and the inconsistent conclusions deduced from 

 them by MM. Laplace and Poisson, I have had occasion to 

 speak in the first volume of this Journal, where I have shewn 

 that, whatever be the ratio of the specific heat of air under a 

 constant pressure, to its specific heat under a constant volume ; 

 if that ratio only be constant, the variations of the quantity of 

 heat in a mass of air must be uniform, while those of its volume, 

 under a constant pressure, form a geometrical progression ; and 

 it is remarkable, that our first-rate authorities on the subject, 

 who admit the constancy of this ratio, did not see that it was di- 

 rectly at variance with the commonly received theory of the air- 

 thermometer. 



But, although the value of the ratio referred to have nothing 

 to do with the true law of temperature, yet its exact determina- 

 tion would be of great moment in various researches. Consi- 

 derable deference is due to the experiments of the illustrious 

 philosophers above mentioned. They were well calculated for 

 shewing that the ratio of the specific heats is constant; because, 

 supposing any inaccuracy to attach to them, it would be com- 

 mon to all the cases. But I had always some doubt whether 

 their apparatus was the most ehgible for determining the exact 

 value of that ratio. The apparatus mostly employed consisted 

 principally of a glass balloon, to the neck of which was fitted a 

 brass cap and stop-cock. From the side of the cap, proceeded a 

 horizontal pipe, communicating with a vertical glass tube, ter- 

 minating in some light liquid to act the part of a very sensible 

 gauge or measure of the variations of pressure. The same ho. 



