78 SECTION PHYSICS. 



forces a reed and a column of air which acts upon it. The column 

 of air restrains, steadies, and purifies the vibrator, which would other- 

 wise be coarse, and not worth hearing. The theory may be summed 

 up in a few words, that an organ pipe is a spring cushion re-acting 

 upon a lamina or plate. When we have a reed pipe, that is easily 

 understood. Here is the reed, and here is the spring cushion of air, 

 its elasticity depending on its bulk and proportion, but when you 

 come to a flue pipe, in which the reed is not apparent, it is not so 

 obvious ; but Mr. Smith regards the reed as still existing, and con- 

 siders that the sheet of air which passes from the mouth of the pipe 

 acts in fact as a reed. Bearing this fully in mind, its application to 

 the reed is easily seen. It one day occurred to me to make a double 

 ring of wire, which I did ; and fixed the two ends into a board, bored 

 a hole in it, and applied an open reed, and I found there was the 

 same result as with the string. We then determined we would analyze 

 what were the real functions of the string and of the reed, and a long 

 series of experiments took place, and at length it was found that in 

 order to gain the asolian tone there were three things necessary : there 

 must be constraint ; there must be sympathetic resistance ; and there 

 must be transmission, in order to gain power ; and those were all 

 contained in the first object I had tried, namely, the ring. The ring 

 first adopted was that of the simple form of a watch-spring, namely, a 

 double ring ; but one day on looking at Lissajous's diagrams, showing 

 the different forms of vibration, an idea occurred to me that a different 

 tone might be gained by a different form of spring. At first I had 

 only a ring, and this afforded perfect constraining effect, but there 

 was still the danger that to get perfect purity of tone you had to go 

 on increasing the power of the ring, until there was so much resist- 

 ance that the reed would not sound. The question was, how to get 

 fulness of tone by getting that ring of a certain size and a certain 

 elasticity, and yet that you should have the power of changing the 

 quality of tone, without altering those favourable conditions? Accord- 

 ingly, I have here eight different type forms of constraints upon the 

 vibrator. Supposing I have a piece of wire, I begin with the frame 

 of the reed, turn it over, come round, and attach the reed to it, and 

 continue its course until it conies round on the other side of the frame, 

 so that it forms a circle. That gave perfect resistance, and a perfectly 



