REFLEXION AND REFRACTION. " 29 



since the phenomena of the colours of thin plates phenomena 

 which first suggested it, to the mind of Newton have been 

 shown to be irreconcilable with it. The explanation which 

 it yields of the facts now under consideration is alike un- 

 satisfactory. In fact, the molecules which are transmitted 

 are not all in the maximum of the fit of transmission ; but are 

 supposed to reach the surface in every possible phase of this, 

 which may be called the positive fit. But as a change of the 

 fit from positive to negative is, in general, sufficient to over- 

 come altogether the effect of the attractive force, and subject 

 the molecule to the repulsive one, it is obvious that the phase 

 of the fit must modify the effects of these forces in every inter- 

 mediate degree ; and that the molecules which do obey the 

 attractive force must have their velocities augmented in 

 different degrees, depending on their phase. Hence, as the 

 direction of the refracted ray depends on its velocity, the 

 transmitted beam should consist of rays refracted in widely 

 different angles, and should be scattered and irregular. 



(36) Let us now turn to the account which the other 

 theory gives of the same phenomena, and of their laws. 



The velocity of propagation, in the wave-theory of Light, 

 depends on the elasticity of the vibrating medium as com- 

 pared with its density. In the same homogeneous medium the 

 velocity will be therefore constant, and the wave propagated 

 from any centre of disturbance spherical. But when a wave 

 reaches the surface of a new medium whose elasticity is diffe- 

 rent, it will give rise to two waves, one in each medium, and 

 both differing in position from the original wave. For it is 

 obvious that, in general, the several portions of the incident 

 wave will reach the bounding surface at different moments of 

 time. Each of these portions will be the centre of two new 

 waves, one of which will be propagated in the first medium 

 with the original velocity, while the other will be propagated 



