REFRACTIVE INDEX 79 



we have a definite quantity to deal with, which only 

 depends on the state of the body, here the nature of the 

 light plays a part also, since the velocity of light depends 

 on its colour, i.e. on its wave-length. It is only in a 

 vacuum that there is no difference in this respect ; the 

 partial occupation of space by matter causes a fall in 

 velocity which is different for different colours. This 

 varying influence causes the phenomenon of dispersion, 

 i. e. the separation of light of different colours, or formation 

 of a spectrum, and the first question is to express the 

 effect exerted on different kinds of light in an intelligible 

 manner, by representing the refractive index as a function 

 of the wave-length (A). This has not yet been accomplished, 

 since the formula of Cauchy : 



B G 



n = A -\ h p > 



serves only for bodies of normal dispersion, whilst the 

 newer formulae which take abnormal, dispersion into 

 account, do not fit the observations with the certainty 

 desirable. We are therefore driven to avoid this difficulty 

 by dealing only with the relations discoverable by using 

 light of a single wave-length l . Measurements refer mostly 

 to sodium light n^. 



C. Effect of Density. 



Not only the wave-length of light, but also the condition 

 of the body in question influences the refractive index. It 

 has been found that the influencing cause is the density 

 of the substance rather than its temperature or state of 

 aggregation, which is in accordance with the fundamental 

 point mentioned on p. 78, that it is the occupation of 

 space by matter that affects the velocity by light. The 

 influence can be traced numerically by the method of 

 Lorenz and Lorentz, the former 2 starting with the as- 



1 Graham-Otto-Landolt, 1898, i, iii. 2. 576. 



2 Wied. Ann. n. 70. 



