24 Prof. J. Le Conte on the Discrepancy between the Computed 



expended in expanding the column of air, or, which is the same 

 thing, in lifting the pressure of the atmosphere on one square 

 foot through a height of one foot, while the remainder remains 

 free and augments the temperature of the mass of air. But, 

 according to the experiments of J. P. Joule*, the quantity of heat 

 required to increase the temperature of one pound of water one 

 degree Centigrade is equivalent to that expended in lifting ] 389*6 

 lbs. through a space of one foot. Hence the amount of heat 

 consumed in lifting 2116-268 lbs. (the atmospheric pressure on 

 one square foot) through a height of one foot would augment the 



21 1 fi*2fi8 



temperature of , OC) ~ = 1'522933 lbs. of water one degree 

 IooU'd 



Centigrade, being the amount of heat expended in expansion. It 



is obvious that this last, subtracted from the whole amount of heat 



employed, gives the quantity of heat which would be required to 



raise the temperature of the same mass of air from 0° to 1° C, 



in case it were prevented from expanding, or kept under constant 



volume. Consequently we have the following numerical data in 



relation to the column of air under consideration : — 



Heat expended under constant pressure . . =5*233074 

 Heat expended in expansion =1*522933 



Hence, heat required under constant volume =3*710141 

 Hence, adopting Laplace's notation, we have 



. c 5-233074 1 0*2379 



c 7 " 3*710141 ~ 0*70898 0*168666 ~^ * 1U *'°- 



Substituting this value in Laplace's formula, we obtain 

 v = 332*4907 metres per second, which, considering the delicacy 

 of the experiments from which the physical data have been drawn, 

 is as perfect an accordance with experiment as the most hyper- 

 critical could demand. In illustration of this fact, it is only 

 necessary to notice that, if the coefficient of dilatation of air be 

 assumed =0*00366666 instead of 0*00367 (the other elements 

 of the computation remaining the same), the value of k comes 

 out =1*40995, and z; = 332*4288 metres per secondf. 



In the foregoing reasoning I have avoided all theoretical con- 

 siderations and other elements entering into the formulae which 

 have been deduced from the " Mechanical Theory of Heat," and 



* Phil. Trans for 1850, p. £3. 



t In the above computation I have taken the specific heat of air =0*2379. 

 Some physicists use the number - 2377 ; while M. Jamin (vide Cours de 

 Physique, vol. ii. p. 370, Paris. 1859), who appears to have had access to 

 the details of Regnault's experiments, gives the number 0*23741. The 



