PAPERS PRESENTED AT GENERAL SESSIONS 25 



particle receives one push in the right direction. It is 

 this one-sidedness of the action which enconrages the os- 

 cillations. Moreover, as the heat here has the tendency 

 to increase the pressure near the node, the oscillations 

 will start very readly. A slight modification of this exper- 

 ment is the glass blower's bnlb, 6, which emits a sound 

 when heated round about the neck, p. Instead of making 

 the lower part of the tube narrower, as in 5, we might 

 proceed as in 7, where the annular area takes the place 

 of the narrow tube, in 6. A modification of this tube is 

 the tube of Fig. 8, which will sound when heated at p, 

 and which is much more sensitive. It is evident from the 

 explanation that this tube ^^'ill not sing when the lower 

 end of the inner tube is open, because the one-sidedness 

 of the action is destroyed. Slight modifications of 8 are 

 the tubes represented by 9 and 10. If we place a hot wire 

 net inside the tubes 5, 6, 7, 8, 9, 10, where the hot flame 

 was out-side, the tubes will produce a sound. In all cases, 

 in the organ pipe, in Rijke's experiment (tube 2) and in 

 the tubes 5-10, the oscillations of a column of air are 

 maintained by a one-sided addition of momentum at the 

 right moment and in the right place. 



These experiments belong to a large variety of phe- 

 nomena in which a direct motion is transformed into a 

 periodic motion, or, electrically speaking, where direct 

 current is transformed into an alternating current. 



