THE PHONOGRAPH, OR VOICE-RECORDER. 293 



frozen during severe cold melting into speech again, so that 

 all the babble of a past day came floating about his ears, 

 has been realized by Edison's invention. Although such 

 expressions may not be, and in point of fact are not, strictly 

 scientific, I am not disposed, for my own part, to cavil with 

 them. If they could by any possibility be taken au pied de 

 la lettre (and, by the way, we find quite a new meaning foi 

 this expression in the light of what is now known about 

 vowels and consonants), there would be valid objection to 

 their use. But, as no one supposes that Edison's phono- 

 graph really crystallizes words or freezes sounds, it seems 

 hypercritical to denounce such expressions as the critic 

 of the Telegraphic Journal has denounced them. 



To return to Edison's instrument. 



Having obtained a material record of sounds, vocal or 

 otherwise, it remains that a contrivance should be adopted 

 for making this record reproduce the sounds by which it was 

 itself formed. This is effected by a third portion of the 

 apparatus, the transmitter. This is a conical drum, or rather 

 a drum shaped like a frustum of a cone, having its larger 

 end open, the smaller which is about two inches in 

 diameter being covered with paper stretched tight like the 

 parchment of a drum-head. In front of this diaphragm is a 

 light flat steel spring, held vertically, and ending in a blunt 

 steel point, which projects from it and corresponds precisely 

 with that on the diaphragm of the receiver. The spring is 

 connected with the paper diaphragm by a silken thread, just 

 sufficiently in tension to cause the outer face of the 

 diaphragm to be slightly convex. Having removed the 

 receiving apparatus from the cylinder and set the cylinder 

 back to its original position, the transmitting apparatus is 

 brought up to the cylinder until the steel point just rests, 

 without pressure, in the first indentation made in the tinfoil 

 by the point of the receiver. If now the handle is turned at 

 the same speed as when the message was being recorded, 

 the steel point will follow the line of impression, and will 

 vibrate in periods corresponding to the impressions which 



