SCIENCE. 



3" 



the terminal that is free ; not because it makes a ground, 

 as it is termed in electrical science, or completes an elec- 

 trical circuit, for if the individual listening be as perfectly 

 insulated as glass or hard rubber can make him, the sound 

 is as loud as if he stood on the ground ; but the individual 

 becomes electrified by induction, it is the same as enlarg- 

 ing the terminal would be. Consequently receivers are 



FlGl'RE 3. 



made having only one wire terminal (see Fig. 3), the 

 other plate being connected by a conductor to a metallic 

 ring upon the knob, and this receiver is as efficient as the 

 other. 



Electricians will recognize in this structure what is 

 technically known as the air condenser, and the mutual 

 attraction of the two plates has been employed as a means 

 of measuring electrical potential. In this case one of the 

 plates is suspended from one arm of a balance, while the 

 other is fixed underneath it at a short distance. The at- 

 traction of the plates when they are electrified requires an 

 extra weight to keep them apart, and the weight needed 

 is the measure of the attractive force. But the plates will 

 attract each other when glass or mica or any other non- 

 conducting substance is placed between them in the 

 place of the air ; and one might expect-that if such an air 

 condenser would give sonorous results, other forms of 

 condensers, would do so likewise, and this is so. Indeed, 

 whoever has charged a Leyden jar has probably noticed 

 the sounds coming from it when it is nearly saturated. 

 In 1863 SirWm. Thompson had his attention directed to 

 the sounds produced by discharge in an air condenser.* 



When the two plates of Epinus's condenser are in met- 

 allic contact no sounds whatever can be produced by it, 

 but if they are separated by a thin film of air they will 



Figure 4. 



Sec papers on Electro-Statics and Magnetism, page 236. 



reproduce speech (see Fig. 6, at E). In the first case the 

 electricity passes from one plate to the other without 

 doing work or changing its form ; while in the latter, its 

 form is changed and work is done, and between the best 

 conductors, such as silver and copper and the perfect 

 non-conductor air there are all degrees of conductibility, 

 and whenever electricity spends its energy upon an im- 

 perfect conductor it results in heating it; that is, in 

 molecular and atomic vibrations. Consequently an un- 

 dulatory current from an ordinary transmitter, when sent 

 through an imperfect conductor, will set up sound vibra- 

 tions in it which may be appreciated by the ear. Let, 

 then, any poor conductor, like a disk of carbon, a sheet 

 of paper or of gelatin, or such chemical substances as 

 ammonium chloride, be placed between the terminal 

 plates, and an undulatory current sent through them will 

 result in sound, and speech may be reproduced. 



Now, the phenomena observed in Geissler's tubes and 

 Crooke's tubes show that the residual gaseous molecules 

 are violently impelled from the electrified terminals, not 

 simply because they are electrified, but because they are 

 heated, for the same phenomena are witnessed when the 

 terminals are heated in other ways ; so it is probable that 

 between the plates of the air condenser there is an actual 

 impulsion of the air particles from one to the other, and 

 that the phenomenon of attraction is not isolated from 

 molecular impact. Receivers have been made in which 

 a vacuum could be produced between the plates, but no 

 great difference could be observed in their performance ; 

 and when one reflects upon the immense number of 



Figure 5. 



molecules left in the best vacuum yet produced, it is not 

 a matter for much surprise. 



When a non-conductor, such as air, or vulcanite, or 

 mica, separates the two plates, there is a complete trans- 

 formation ol the electricity at the limiting surfaces, and 

 with small condensers the efficiency depends upon the 

 electromotive force employed. For low electromotive 

 forces, such as common batteries of a few cells can give, 

 the effect is almost inappreciable, and for this reason 

 such a receiver as this is quite free from the disturbance 

 known as induction, and which is so troublesome in the 

 magneto-telephone, such induced currents being generally 

 of low electromotive force. 



Among the earliest of my experiments, made while de- 

 veloping this method, was to attach one terminal wire 

 from an induction coil to the outer coating of a Leyden 

 jar, taking the other wire from the coil in one hand, and 



