200 



ME. W. F. G. SWANN ON THE SPECIFIC HEATS OF AIR AND 



Section Page 



(24) The specific heat of carbon dioxide 232 



(25) Calibration of the fine tubes for carbon dioxide 233 



(26) Abridged tables for the calibration of the fine tubes for carbon dioxide 234 



(27) The calorimetric experiments on carbon dioxide 234 



(28) Abridged tables for the main calorimetric experiments on carbon dioxide 235 



(29) Variation of the specific heat at constant pressure with pressure 237 



(30) Corrections to be applied to the final values 237 



(31) Comparison of the results for carbon dioxide with the theoretical value 237 



(L) INTRODUCTION. The first reliable investigation of the specific heats of gases was 

 made by RE<;NAULT,* who in an elaborate series of experiments measured the specific 

 heats of several gases at constant pressure by the method of mixtures. This method, 

 which in modified forms has been employed by nearly all subsequent observers, 

 presents great difficulties when applied to gases, and further, it only gives the mean 

 values of the specific heats over large ranges of temperature, and it is consequently 

 not well adapted to the detection of small variations of the specific heat with 

 temperature. During the period in which I was at the Royal College of Science, 

 London, Prof. CALLENDAR suggested that I should make some measurements of the 

 specific heats of air and carbon dioxide by the continuous electric method, which is 

 particularly well adapted to the measurements of specific heats over small ranges of 

 temperature. I should like to take this opportunity of expressing my deepest thanks 

 to Prof. CALLENDAR for his advice and encouragement throughout the work. 



(2) Outline of the Method. A constant stream of gas is passed through a tube 

 GF, fig. 1, to be presently described, into a jacket J, which surrounds a central tube 



AB. The gas passes into the tube AB at the 

 JL_ P temperature of the jacket, and is there heated by an 

 |ii' F T^ electric current passing through a platinum coil P, 



the rise in temperature being measured by two 

 platinum thermometers, K and H, used differentially. 



Pig- 1 If C is the electric current, E the potential 



difference between the ends of the heating coil, Q 

 the mass of gas which flows through the apparatus 

 per second, $0 the rise in temperature of the gas, J 

 r= the mechanical equivalent of heat, and S the specific 

 heat of the gas at constant pressure, then neglecting 

 for the present the heating effect of the leads of the 



heating coil, we have, when the temperature conditions of the different paxts of the 

 apparatus have become steady, 



CE = JSQS0+AS0, (1) 



where h 80 is a term representing the heat loss. If now the rate of flow of the gas is 

 * ' Memoires de L'Academie des Sciences de L'Institut Imperial de France,' 1862. 



