278 LETTERS ON NATURAL MAGIC. 



which are very indistinctly visible, and have a tremulous 

 motion, as if they were "dancing in the air." The very 

 same effect is perceived when we look at objects through 

 spirits and water that are not perfectly mixed, or when 

 we view distant objects over a red-hot poker or over a 

 flame. In all these c-ises the light suffers refraction in 

 passing from a medium of one density into a medium of 

 a different density, and the refracted rays are constantly 

 changing their direction as the different currents rise in 

 succession. Analogous effects are produced when sound 

 passes through a mixed medium, whether it consists of 

 two different mediums, or of one medium where portions 

 of it have different densities. As sound moves with 

 different velocities through media of different densities, 

 the wave which produces the sound will be partly reflected 

 in passing from one medium to the other, and the direc- 

 tion of the transmitted wave changed; and hence in 

 passing through such media different portions of the wave 

 will reach the ear at different times, and thus destroy the 

 sharpness and distinctness of the sound. This may be 

 proved by many striking facts. If we put a bell in a 

 receiver containing a mixture of hydrogen gas and atmo- 

 spheric air, the sound of the bell can scarcely be heard. 

 During a shower of rain or of snow, noises are greatly 

 deadened, and when sound is transmitted along an iron 

 wire or an iron pipe of sufficient length, we actually hear 

 two sounds, one transmitted more rapidly through the 

 solid, and the other more slowly through the air. The 

 same property is well illustrated by an elegant and easily- 

 repeated experiment of Chladni's. When sparkling 

 champagne is poured into a tall glass till it is half full, 

 the glass loses its power of ringing by a stroke upon its 

 edge, and emits only a disagreeable and puffy sound. 

 This effect will continue while the wine is filled with 

 bubbles of air, or as long as the effervescence lasts ; but 



