M. A. Kundt on the Velocity of Sound in Tubes. 25 



exchange of heat between air and wall, it is clear that the velocity 

 of sound, if an exchange of heat everywhere takes place, must be 

 the smaller the narrower the tube, the rougher its wall, or the 

 larger and rougher the surface of a second tube inserted. In 

 like manner powder must have greater influence the finer it is 

 and the narrower the tube ; for it is then better distributed in the 

 sounding air*. 



The greater, finally, the duration of the oscillation of a tone, or 

 its wave-length, the more considerable is the exchange in the 

 longer time; hence the diminution of velocity must, under the 

 same circumstances, be greater. 

 X Since, as has been said, all the influences of the wall, of the 

 powder, and so forth are infinitely small in their action on the 

 velocities when the diameter of the tube reaches a certain mag- 

 nitude, the method, notwithstanding the discrepancies observed, 

 may be used for exact determinations of wave-lengths, and there- 

 fore of velocities of sound, if the tubes for the dust-waves are wide 

 enough. The author had proposed, in the next place, to test 

 more extensively than was hitherto possible two deductions from 

 the theory — that is, that the velocity of sound under different 

 pressures is the same, and that its variation by the temperature 

 is expressed by the factor V\ -\-ctt. 



The velocity of sound in air was determined in wide tubes be- 

 tween pressures of 400 and 1760 millims. Within these limits 

 of pressure, which correspond to a difference in pressure of nearly 

 two atmospheres, the velocity of sound remained quite constant. 



To ascertain whether the influence of the diameter of the tubes, 

 when narrower ones are used, is the same under different pres- 

 sures, the velocity was determined in some narrower tubes under 

 a pressure of from \ to 2^- atmospheres. Using a note whose 

 semiwave-length was 45 millims. in a tube of 3*5 millims. dia- 

 meter, the velocity of sound under a pressure of 2^ atmospheres 

 was about 4 metres greater than under a pressure of half an 

 atmosphere. 



The author refers this to the circumstance that, with air at 

 greater density, the greater quantity of heat produced and con- 

 sumed is not exchanged with the wall in the same ratio as the 

 smaller quantity of heat with a smaller density. 



The velocity of sound at 0° and 100° was then determined, 

 and in tubes which were so wide that their walls had no influence. 

 One side of the apparatus with one dust-wave tube being im- 

 mersed in melting ice, the other with its dust-wave tube was 



* Powder, by its increasing the mass to be moved, might somewhat im- 

 pede the velocity. I have convinced myself, however, by experiments 

 which will be discussed in the complete paper, that this circumstance is in- 

 adequate for the explanation of the considerable variations. 



