2IO Of the Light manifested in Combustion. 



tain that it ought to be found pre-existing in some of 

 the bodies that are decomposed in that operation ; and 

 there is every reason to suppose, if that were really the 

 case, that the quantity of light disengaged in the com- 

 plete combustion of a given quantity of any given 

 inflammable substance would be limited, and just as 

 invariable as all the other chemical products of that 

 process. 



But if light be not a substance emitted by luminous 

 bodies, but a vibration and undulation in an ethereal 

 fluid, analogous to the vibration and undulation of 

 the air which is the immediate cause of sound (as 

 many distinguished philosophers have supposed), in that 

 case we ought to search for the cause of the light 

 which is diffused by the flame of a burning body in 

 the very high temperature of the particles of matter 

 which compose that flame. These particles must be 

 considered as being luminous, in consequence of the 

 action of the same cause which renders a cannon 

 bullet luminous which has been heated red-hot in the 

 fire. And as all known bodies cease to shine in the 

 dark at a known given temperature (that of about 

 1000° of Fahrenheit's scale), the hot particles which 

 compose a visible flame ought to disappear entirely 

 the moment they become cooled down to that tem- 

 perature. 



If we adopt this hypothesis respecting light (which 

 I confess has ever appeared to me to be the most prob- 

 able), we must no longer expect to find the quantities 

 of light excited in the process of combustion to be in 

 any constant ratio to the quantities of inflammable 

 matter burned : so far from it, we should be obliged to 

 admit that the discovery of such an invariable relation 



