CARBON DIOXIDE AT DIFFERENT TEMPERATURES. 231 



volume and at zero pressure, and if S,, and s refer to the same temperature, 

 S *o = R = 2'8727xl0 8 C.G.S. units, or 0'06872 calorie. JOLY has measured the 

 specific heat of air at constant volume by the steam calorimeter.* In his calcu- 

 lations, however, he assumes REGNAULT'S value for the latent heat of steam 

 (L = 5367 cal./gr.). This value is almost certainly too low. HENNING has recently 

 measured the latent heat of steam t and found 538 - 9 caL/gr., while JOLY'S own 

 observations with the steam calorimeter lead to the value 540 '2 in terms of the 

 calorie at 20 C. Assuming this value to correct JOLY'S results for s, and extra- 

 polating to zero pressure, we obtain for the value of s a at 50 C. (the temperature at 

 which JOLY measured it) s = 0'1732. This value of s , in conjunction with my own 

 value of S at 50 C., gives - OG87 calorie for S , which agrees with the above 

 value of R to less than TFOO of the specific heat. 



(23) Criticism of Previous Results. All previous determinations of the specific 

 heat of air at constant pressure have given a lower value than 0'2422 (my own value 

 at 50 C.). REGNAULT obtained - 2375, which is lower than 0'2422 by about 2 per 

 cent. One has only to read REGNAULT'S paper to feel convinced that an error of 

 2 per cent, could not have arisen in his experiments from lack of care in the 

 observations. If my own value is correct, it is to some point in the theory of 

 REGNAULT'S method that we must look if we are to explain an error of 2 per cent. 



Now, in order to determine the alteration A# in the temperature of the calorimeter 

 per minute due to radiation, conduction, &c., REGNAULT assumed an equation of the 

 form A# = A B#, where 9 is the excess of the temperature of the calorimeter over 

 that of the room, B$ represents the radiation loss per minute from the calorimeter, 

 and A represents the alteration in the temperature of the calorimeter per minute due 

 to conduction of heat from the heating bath to the calorimeter along the pipe 

 connecting them. The total alteration in temperature, due to the above causes during 

 the experiment, was obtained by summation. The constants A and B were determined 

 from two equations obtained by observing the alteration in temperature of the 

 calorimeter before and after the main experiment. Now, since there was no gas 

 flowing through the connecting pipe when the observations which determined the 

 constants A and B were made, the conduction effect must have been very different 

 from what it was when gas was flowing through the pipe. 



To fix our ideas, suppose the temperature gradient in a connecting pipe is uniform 

 when no gas is flowing through it. Now, when the hot gas passes through the tube, 

 it will keep the temperature of the latter up, so that the temperature gradient in the 

 tube near the heater will be very much reduced ; in fact, it is possible that it may be 

 reduced practically to zero, so that no heat will be conducted from the heater to a 

 point In the pipe a little distance from it. Heat will certainly be conducted to the 

 calorimeter through the pipe, but this heat will come from the gas itself and not 



* ' Phil. Trans.,' 1904. 



t ' Annalen der Pbysik,' No. 10, 1904. 



