PHILOSOPHICAL SOCIETY OF WASHINGTON. 149 



be formed among the particles, and the outside air would then be 

 absorbed, as water is by a sponge when the pressure of the hand is 

 removed. 



I imagine that in some such manDer as this a wave of conden- 

 sation is started in the atmosphere each time a beam of sunlight 

 falls upon lamp-black, and a wave of rarefaction is originated when 

 the light is cut off. We can thus understand how it is that a substance 

 like lamp-black produces intense sonorous vibrations in the surrounding 

 air, while at the same time it communicates a very feeble vibration to 

 the diaphragm or solid bed upon which it rests. 



This curious fact was independently observed in England by Mr. 

 Preece, and it led him to question whether, in our experiments with 

 thin diaphragms, the sound heard was due to the vibration of the 

 disk or (as Prof. Hughes had suggested) to the expansion and con- 

 traction of the air in contact with the disk confined in the cavity 

 behind the diaphragm. In his paper read before the Royal Society 

 on the 10th of March, Mr. Preece describes experiments from 

 which he claims to have proved that the effects are wholly due to 

 the vibrations of the confined air, and that the disks do not vibrate 

 at all. 



I shall briefly state my reasons for disagreeing with him in this 

 conclusion : 



1. When an intermittent beam of sunlight is focussed upon a 

 sheet of hard rubber or other material, a musical tone can be heard, 

 not only by placing the ear immediately behind the part receiving 

 the beam, but by placing it against any portion of the sheet, even 

 though this may be a foot or more from the place acted upon by 

 the light. 



2. When the beam is thrown upon the diaphragm of a " Blake 

 Transmitter," a loud musical tone is procuced by a telephone con- 

 nected in the same galvanic circuit with the carbon button, (A,) 

 Fig. 4. Good effects are also produced when the carbon button (A) 

 forms, with the battery, (B,) a portion of the primary circuit of 

 an induction coil, the telephone (C) being placed in the secondary 

 circuit. 



In these cases the wooden box and mouth-piece of the trans- 

 mitter should be removed, so that no air-cavities may be left on 

 either side of the diaphragm. 



It is evident, therefore, that in the case of thin disks a real vibration 

 of the diaphragm is caxised by the action of the intermittent beam, in. 



