[ 34 ] 



V. On the Velocity of Sound. 

 By Dr. H. W. Schroder van der Kolk*. 



ALTHOUGH the determination of the velocity of sound is 

 in itself a problem of great interest, the exact knowledge 

 of this constant has become doubly interesting since the obser- 

 vation that it was fitted to furnish one of the most accurate 

 methods of determining the mechanical equivalent of heat. The 

 attentive examination of this problem is therefore a matter of the 

 greatest interest. 



The methods that have been employed to measure the velocity 

 of sound may be divided into direct and indirect. 



To the latter class belongs Wertheim and Masson's method 

 by means of organ-pipes, which depends on the observation of 

 the pitch and of the wave-length. The latter, however, is very 

 difficult to determine with precision ; and the difficulty of obtain- 

 ing accurate numerical results is increased by the fact, that the 

 quantity which is directly measured requires to be multiplied by 

 a large number, in order to obtain the distance through which 

 sound travels in a second, and hence the errors of observation, 

 increased in the same proportion, are contained in the result. 

 Accordingly this method has not yielded any accurate results. 



The second indirect method is that of Clement and Desormes, 

 subsequently put in practice also by Gay-Lussac andbyMasson. 

 In this method, as is well known, the magnitude directly fur- 



c 

 nished by experiment is the value of - (where c and c l are the 



c i 

 specific heats under constant pressure and constant volume), and 

 the velocity of sound is calculated from this by means of the 

 well-known formula. This method is susceptible of greater accu- 

 racy, and in the hands of Masson it appears actually to have led 

 to good results. But even these determinations are incapable 

 of yielding a perfectly exact value, inasmuch as, so far as I am 

 aware, the full details of the experiments have not been published. 



Hence there remain for us to consider only the direct methods. 



The method that has been most generally adopted consists in 

 determining the time in which sound traverses a given distance. 

 At one station a signal is given, usually by means of a cannon, 

 and at the other station the time is noted which elapses between 

 the flash of the powder and the arrival of the sound. 



Another method proposed by Dr. Bosschaf is probably sus- 

 ceptible of great accuracy, but hitherto it has not been tried 

 upon a large scale. 



* Translated from Poggendorff's Annalen,\o\. cxxiv. p. 453 (1865, No. 3), 

 by G. C. Foster, B.A. 



t Poggendorff's Annalen, vol. xcii. p. 485. 



