Belas — Water-jets, and the Effect of Sound thereon. 363 



a simple vibration having a frequency of 128. In order to explain 

 the real state of affairs more easily, let me refer once more to 

 Part i. of this subject. 



Plate XIX., No. 2, shows a jet of water broken up by a single 

 blow on the support. There is a long bar of liquid separated, 

 having indentations upon it, which cause it to break into separate 

 drops lower down, on account of its instability. The resolution 

 of a water- jet falling freely is determined by accidental tremors 

 and friction in the pipe, but is rendered rhythmical to a large 

 extent by the vibrations due to the impacts of the drops on the 

 sides of the sink or other vessel reacting on the orifice. The 

 effect of a blow is to cause a single disturbance of large amplitude 

 to travel down the jet, and to break it up considerably nearer the 

 orifice, and so the bar of liquid is formed. But the undulations 

 impressed upon that part of the fluid column still persist, and 

 effect its resolution into smaller drops, after it has been cast off 

 from the main column. 



This is, I believe, similar to what takes place in the jet when the 

 difference-tone is produced. The disturbance of large amplitude 

 caused by the periodic agreement of the phases of the separate 

 forks breaks up the column into long bars, which have the 

 smaller undulations impressed upon them, and these bars are 

 again resolved as they fall. If we remove the two intermediate 

 groups of three drops from fig. in., as has been done in fig. iv., 

 we have left three drops separated from each other by a space of 

 6 cms. 



Furthermore, the first and third are oblate spheroids, while 

 that in the centre is prolate, showing that a space corresponding 

 to at least a complete period of oscillation of the drop has been 

 included. 



The accepted theory of combinational tones is that due to 

 Helmholtz; and it depends upon the fact that when the 

 amplitudes of the generating tones are large, the restoring force 

 can no longer be considered proportional simply to the displace- 

 ment, but also to its square. Now such combinational tones 

 (produced by the double siren, or harmonium reed, or two 

 singing-flames) are only heard when the generating tones are 

 very strong, and then as present along with them. 



But in this instance, the generating tones disappear, and the 



