246 ON LIGHT. 



internal refraction at that of the second, or the "relative 

 index of refraction," is constant for the same media, and 

 is equal to the quotient of their respective absolute re- 

 fractive indices. Thus, if the first medium be water, 

 and the second be plate-glass, whose respective absolute 

 indices are f and I, the relative index, or that out of 

 water into glass, will be \ divided by I- or 1=^1 -125. 



(30.) A very curious result follows from what has been 

 said, viz., that though light can pass out of a rarer 

 medium into a denser, whatever be the obliquity of 

 incidence, even when the incident ray but just as it 

 were grazes the surface, yet the converse is not the case. 

 For every denser medium, there is a limit of obliquity, 

 beyond which transmission into a rarer cannot take 

 place. The ray is wJwlIy reflected without undergoing 

 any diminution of brightness whatever; observing the 

 same law of equality between the angles of incidence 

 and reflexion, as in the case of ordinary reflexion on a 

 mirror. The brightness of the reflexion, however, far 

 surpasses anything that can be obtained from the most 

 brilliant looking-glass or metallic mirror, being equal to 

 that of the object directly seen. The eff"ect is very 

 striking, and is easily seen by immersing a small rod 

 obliquely in a glass tumbler of water, and viewing the 

 under surface of the water from below upwards at a 

 moderate obliquity. The reflexion of the rod is seen 

 without the smallest diminution of brightness. It ^s 

 thus tliat fishes see the bottom of their pond redupli- 

 cated by internal reflexion on the distant parts of its 

 surface. The rationale is simple enough. If two angles 



