272 Mr. 11. H. M. Bosanquet on the Theory of Sound. 



placed before the readers of the ' Philosophical Magazine/ 

 These I propose to submit to you. 



Yours &c, 



R. H. M. Bosanquet. 



1. The Velocity of Sound, and Medio of Specific Heats, in Air. 



On examining the received values of the velocity of sound, 

 it becomes apparent that the discordance between the results 

 of different experimenters is considerable. I propose to 

 put the results together and compare the different experi- 

 mental data which can be brought to bear on the subject. 

 Without attempting to give a final result, I think some con- 

 siderations of interest can be suggested as to the values that 

 most nearly satisfy the different determinations and require- 

 ments. 



The following Table represents the velocities of sound found 

 by different observers, reaching from the middle of the last 

 century up to the present time : — 



Names of observers. Velocity of sound. 



metres. 

 Academie des Sciences (1738) 332 



{333*7 



Goldingham (1821) 331-1 



Bureau des Longitudes (1822) 330*6 



Moll and Van Beek 332*2 



Stampfer and Myrbach 332*4 



Bravais and Martins (1844) 332*4 



Wertheim 331*6 



Stone 332*4 



Le Roux.... 330*66 



Regnault 330*7 



It is clear that we cannot come to a final conclusion on these 

 premises as to the number to be adopted. I do not think 

 it is a case where the method of least squares is applicable : 

 for the differences appear to be systematic ; and the question 

 is, which observers do we really trust most ? 



The difficulty of the application of the method of discharges 

 of cannon to the determination of an element, in the theory of 

 which all motion is supposed continuous and all displacements 

 small, has been frequently noticed. The exhaustive researches 

 of Regnault* on the propagation of disturbances in tubes, in 

 which they were examined at different periods of their course, 



* Mem. de I'Institut, 1868. 



