274 Mr. R. H. M. Bosanquet on the Theory of Sound. 



gated backward and forward through a tube '07 metre in 

 diameter, which should diminish the velocity. 



As between Regnault's results and those of any one else, I 

 have no doubt that in Regnault's we know far more about the 

 different elements of the question in a trustworthy manner 

 than in any other case. But it does not follow that his num- 

 ber is to be taken as true to the first place of decimals. 



If we examine the isolated determinations, we find that the 

 deviations are not inconsiderable : in some cases they amount 

 to fully the difference between Regnault's own number and 

 332 metres, or thereabouts, which is more usually accepted. 



Under these circumstances we can form no definite con- 

 clusion from the numbers before us ; but I think that if we 

 take Regnault's number (330*7 metres) and the ordinary 

 number (332 metres), the mean between them, 331*35 metres, 

 or thereabouts, will afford a number which will be safe to work 

 with, and may be used with a certain reserve as representing 

 the results of experiment. 



There is a point in the kinetic theory of gases upon which 

 this number has a bearing of great interest, as every body 

 knows. It may be that the particular detail I am about to 

 mention is well known ; but I have never come across any allu- 

 sion to it. 



The connexion between the ratio of the specific heats and 

 the observed velocities of sound is as follows (I insert the 

 velocities in feet here, as we are generally here accustomed to 

 reckon this particular datum in this way) : — 



atio. 



• Velocity of sound. 







feet. 



metres. 



•41 



1090-7 



332-4 



•40 



1086-8 



331-2 



•395 



1085-0 



330-6 



These numbers are reckoned with data which give 918'5 

 feet for the Newtonian velocity. 



Regnault, selecting the velocity which he applies to the 

 specific heats, and applying small corrections, gets 1*3945 for 

 the ratio he adopts. 



It has been long recognized that a relation exists between 

 the nature of the particles of a gas and its ratio of its specific 

 heats. The first essential law is that first proved, I believe, 

 by Clausius*, that the ratio of the vis viva of translation 



to the total vis viva in a gas is § ( 1 J , where — is the 



ratio of specific heats. And the second is that given by Max- 

 * Pogg. Ann. 1857, vol. c. p. 353 {Abhandluncjen &c. 2te Abth. p. 258). 



