166 



THE ROYAL SOCIETY OF CANADA 



nl nl 



- or — = fmr 



The amplitude becomes very great when sin 

 where m has all integral values. Or since c = N\ and N 



c c 



n 



2^ 



resonance occurs when l = \m\ where m is integral. The reflector 

 must, therefore, be placed in the tube to accord with this condition. 



As in Kundt's method, the end of the bar is the exciting source 

 for the air waves, and therefore the same method of measurement 

 regarding the vibration of the bar from the vibration of the air 

 column will apply. In other words if the stopped end be placed so 

 that 1 = ^ m\ the nodes are formed in the dust in the tube at the 



X 

 positions x = o, x= -, x = X, etc., so that the distance between two 



successive nodes is one half the wave length. Hence, recalling that 

 the total length of the bar is one half the wave length in the metal, 

 we may find the velocity of the wave in the metal from the velocity 

 of the wave in the tube by multiplying the latter by the ratio of the 

 length of bar to the distance between nodes in the tube. 



The following table gives a typical set of readings. The bar 

 used was of soft cast steel, 2.54 cms. in diameter and 30 cms. long. 

 The glass tube was 1.4 cms. internal diameter. 



These values give the wave lengths of the air wave in the tube 

 as 11.9 cms. and the frequency of vibration of the bar 8580 per second. 



The velocity of the air waves in the tube which was used here is 

 not the same as the velocity of sound in free air. In consequence, 

 the results obtained by Blaikley^ for the velocity of air waves in a 



iîPhil. Mag. 1884. 



