360 ON LIGHT. 



the other resolved portion. Suppose one end of a long 

 horizontal cord fastened to a wall, the other held in the 

 \iand, and tightly strained. If a small vibratory motion 

 cross-wise to the cord be given to the hand in a hori- 

 zontal plane, an undulation confined to that plane, will 

 run along the cord ; if in a vertical one, then will the 

 undulation be wholly performed in a vertical plane. If 

 the propagation of a wave along a stretched cord be 

 assimilated to that of a ray of light, the former of 

 these cases will convey the idea of a ray polarized in 

 a horizontal, the latter in a vertical plane. If the 

 movement of the hand (always transverse to the cord) 

 be not confined to any particular plane, but take place 

 sometimes in one, sometimes in another, at all sorts 

 of inclinations to the horizon the undulation which 

 runs along the cord in this case will convey the idea of 

 an unpolarized ray. (According to Sir David Brewster*. 

 however, a partially polarized ray would, in this manner, 

 of viewing it, be assimilated to the case when the vibra- 

 tory movement should neither be strictly confined to one* 

 plane, nor altogether irregular, but confined in its devia- 

 tions from it to some angular limit less than a right angle.) 

 (141.) There is nothing to lead us to believe that the 

 vibratory motions of the particles of material bodies, 

 especially those in the state of gases in the act of com- 

 bustion, in virtue of which they are luminous, are ne- 

 cessarily confined to any particular plane. Many thou- 

 sands, or even millions, of vibrations in one plane may 

 be succeeded by as many in any other, according to the 

 direction and frequency of the shocks which Hve rise to 



