[ 32 ] 





IV. On the Velocity of Sound in Tubes. 

 By A. Kundt*. 



TWO years ago the author laid before the Academy a method 

 of determining the velocity of sound in elastic fluids, 

 which consists in setting in vibration the column of air enclosed 

 in a longitudinally vibrating tube. These vibrations may be 

 made visible by light powder strewed in the tube ; and by mea- 

 suring the resultant dust-figures, the wave-lengths of the same 

 exciting tone in different gases, and therewith the corresponding 

 velocities of sound, may be readily foundf. 



Since it is not always certain in these experiments that the ex- 

 citing tone of the longitudinally sounding glass tube remains in 

 successive experiments exactly the same, the apparatus was first 

 of all so altered that not only did one end of the longitudinally 

 vibrating tube produce dust-figures in an enveloping tube, but 

 also the other end in a second tube. If then the tube on one 

 side was filled with the gas to be investigated, and that on the 

 other with air, and the dust-figures were produced, the wave- 

 lengths measured in the gas and in air, belonging to precisely 

 the same tone, gave the exact ratio of the velocities of sound 

 in the gas and in air. 



When, however, the velocities of sound in the simple gases 

 were examined by this method, it was observed that, in spite of 

 the utmost care in purifying and drying the gases, and in spite 

 of frequent endeavours, discrepancies were met with in the deter- 

 minations, which seemed to indicate that sources of error as yet 

 unknown influenced the result. 



It was, moreover, ultimately found that these sources of error 

 are not specially incident to the method used, but under certain 

 circumstances always occur when sound is transmitted through 

 tubes. The author accordingly subjected to a minute examination 

 all the circumstances which might influence the velocities of 

 sound in elastic fluids in tubes. At first, for the sake of simpli- 

 city, only air was used. 



The mode in which the experiments were made was as fol- 

 lows : — The tube on one side of the apparatus was always un- 

 changed, so that the dust- figures were always produced in it 

 under the same conditions, and their length could therefore be 

 taken as the normal length ; on the other side were placed the 

 tubes to be investigated, in which the velocity of sound was to be 

 determined under varied conditions; and the dust-figures were 

 simultaneously produced in the two. The ratio of the wave- 



* Translated from the Berlin Monatsbericht for December 1867. 

 t Phil. Mag. January 1868, p. 41. 



