26 M. A. Kundt on the Velocity of Sound in Tubes. 



heated in steam, and the wave-lengths of the same tone deter- 

 mined at 0° and 100°, which thus gave the ratio of the velocities 

 of sound. 



The following Table gives, as the result of seven experiments, 

 the velocity of sound for 100°, taking the velocity at 0° as 332*8 

 metres : — 



Metres. 



1 388-20 



2 388-84 



3 38915 



4 388*47 



5 389-02 



6 389-64 



7 388-60 



Mean . . 388*99 



Calculating the value from the formula 3328 \/l + 100a 

 (a = 0-003665), we get 389*03. The.difference of the observed 

 and of the calculated value amounts therefore to 0'04 metre. 



It is therefore proved that up to 100° C. the change in the ve- 

 locity of sound is accurately expressed by the factor V\ + ctt, 



c 

 and therefore that Laplace's factor — , the ratio of the specific 



heats of air, is constant within very narrow limits. 



As this cod stau cy is thus experimentally proved, observations 

 of the velocity of sound at 100° may conversely be used to de- 

 termine a, the coefficient of expansion of air. 



Although a is not obtained with excessive accuracy from a 

 single determination, the deviations of a in various observations 

 are not too far apart ; and from the mean of the seven observa- 

 tions we get 



a = 0-003662. 



We see from this that, in spite of the considerable changes in the 

 velocity of sound in tubes, the method in question with a suitable 

 choice of apparatus yields very accurate determinations ; and, in 

 the future, accurate values for the velocities of sound of different 

 gases, and therewith of the ratios of their specific heats, may be 

 expected from it. x^ 



