and the Observed Velocity of Sound in Air and Gases. 25 



have preferred to draw my conclusions exclusively from the phy- 

 sical data furnished by the experiments of Regnault and Joule. 

 Prof. W. J. M. Rankine*, by using the same experimental data, 

 in connexion with certain views in relation to the dynamical 

 theory of heat which are more or less hypothetical, has arrived at 

 nearly the same value of k. He finds it equal to 1*4094; hence 

 the velocity of sound by Laplace's formula would be 332-3636 

 metres per second. It is proper to add that several years ago 

 Mr. Joule himself t deduced the same general result in regard to 

 the verification (which his determination of the mechanical equi- 

 valent of heat furnished) of the truth of Laplace's explanation of 

 the excess of the observed above the calculated velocity of sound. 

 But the imperfect and inaccurate character of the physical data 

 (especially in respect to the specific heat of air) which were at that 

 time at his command did not enable him to submit the question 

 to a severe numerical test. The fact that the development of 

 the " Dynamical Theory of Heat" has cast so much light on the 

 theory of the propagation of sound-waves in the atmosphere, is 

 one among the many illustrations of the interconnexion of all 

 branches of physical science. 



From the discussion of the several points embraced in this 

 paper, the following conclusions seem to be fairly and legitimately 

 deducible; viz. — 



1. That the disengagement of heat (in accordance with the sug- 

 gestion of Laplace) during the propagation of aerial sound-waves 



following 



comparison will show the influence 



on 



the values of k and v : — 





Specific heat of air. 



Value of k. 







Valae of r. 



• 





0-2379 

 0-2.577 

 0-23741 



1-410478 



1-41097 



1-41168 







metres. 

 332-4907 

 332 548 

 332632 





If we as 



foot-pouna 



sume Joule's equivalent of heat to 

 l s for 1 degree Centigrade, the follov 



he 

 ring 



1390, instead < 

 numbers are oh 



rf 1389-6, 

 tained : — 





Specific heat of air. 



Value of k. 







Value of v. 







0-2379 

 0-2377 

 0-23741 



1-4103 

 1-4108 

 1-4115 







metres. 



332-471 



332 528 

 332612 





* Phil. Mas. S. 4. vol. v. p. 485 (1853) ; also Transactions of the Royal 

 Societv of Edinburgh, vol. xx. p. 589. 



t Phil. Mag. S. 3. vol. xxxi. p. 114 (184/). 



