1856] WRITINGS OF JOSEPH HENRY. 417 



slab of marble. This was proved by first determining the 

 length of continuance of the vibrations of a tuning-fork on 

 a thin board, which was afterwards cemented to a flat piece 

 of marble. 



A series of experiments was next commenced with refer- 

 ence to the actual reflection of sound. For this purpose a 

 parabolic mirror was employed, and the sound from a watch 

 received on the mouth of a hearing-trumpet furnished with 

 a tube for each ear. The focus was near the apex of the 

 parabola, and when the watch was suspended at this point 

 it was six inches within the plane of the outer circle of the 

 mirror. In this case the sound was confined at its origin, 

 and prevented from expanding. No conjugate focus was 

 produced, but on the contrary the rays of light, when a 

 candle was introduced, constantly diverged. The ticking of 

 the watch could not be heard at all when the ear was applied 

 to the outside of the mirror, while directly in front it was 

 distinctly heard at the distance of thirty feet, and with the 

 assistance of the ear trumpet at more than double that dis- 

 tance. When the watch was removed from the focus the 

 sound ceased to be audible. This method of experimenting 

 admits of considerable precision, and enables us to directly 

 verify, by means of sound transmitted through air, the 

 results anticipated in the previous experiments. A piece of 

 tissue-paper placed within the mirror and surrounding the 

 watch without touching it, slightly diminished the reflection. 

 A single curtain of flannel produced a somewhat greater 

 efi'ect, though the reflecting power of the metallic parabola 

 was not entirely masked by three thicknesses of flannel ; and 

 I presume very little change would have been perceived had 

 the reflector been lined with flannel glued to the surface of 

 the metal. The sound was also audible at the distance of 

 ten feet when a large felt hat without stifl'ening was inter- 

 posed between the watch and the mirror. Care was taken 

 in these experiments so to surround the watch that no ray 

 of sound could pass directly from it to the reflecting surface. 



With a cylindrical mirror, having a parabolic base, very 

 little increased reflection was perceived. The converging 



27-2 



