266 ON LIGHT, 



no less varied, and in the highest degree characteristic. 

 The presence in the flame of each particular chemiccl 

 ele7?ie7it determines the presence in its Hght of some one 

 or more coloured rays oi definite refrangibility and colour, 

 producing often in its spectrum the appearance of a 

 definite line of coloured light out of all proportion 

 brighter than the rest. Thus the presence of soda in 

 any flaming body is characterized by a narrow and 

 exceedingly vivid line of yellow light. So completely 

 characteristic are these lines of the chemical elements to 

 which they bear relation, that no less than four new 

 metals, Thallium, Rubidium, Caesium, and Indium ow^e 

 their first discovery to the observation of definite spectral 

 lines of their appropriate colour, produced by their 

 presence in quantities too minute to be rendered sen- 

 sible in any other manner.* 



(51.) It is impossible in the compass of a lecture like 

 the present, to do more than notice with extreme brevity 

 these remarkable classes of phcenomena, and that only 

 as bearing upon the general object we have in view. 

 They prove in the most convincing manner the close 

 and intimate relation in which light stands to matter. 

 It enters into the interior of the hardest and least pene- 

 trable bodies, and thereout brings us information of an 



* In reference to what is now called " Spectrum Analysis," in a 

 chemical point of view, I may be here allowed to call attention to a 

 passage in my "Treatise on Light," published in 1827 (Encyc. 

 Metrop., vol. iv. ): "The colours thus communicated by the dif- 

 ferent bases to flame, afford in many cases a ready and neat way of 

 detecting extremely minute quantities of them." Article, " Light," 



524.) 



