2So ON LIGHT. 



necessity that the front oitheir line must to a certain ex 

 tent fall back and make an obtuse angle at its point of 

 junction with that of the unimpeded line. Thus, in our 

 figure, B E will represent the line of division between the 

 two regions, B b the advancing front of the troop when 

 the first man arrives at that line, e e that of the portion 

 still on the good ground after some time elapsed, and 

 EP that of the other portion who have been unable to 

 keep up to the same rate of march. And as the neces- 

 sity of keeping step and not crossing each other's line 

 of march will oblige each man to step out right in from 

 {i.e., at right angles to the new frontage), the progre?^, 

 B p, made by the first man after crossing the line, will be 

 perpendicular to e p, and will be to what he would have 

 made (bm) had it not been for the retardation, in the 

 proportion of his new to his former velocity of march. 



{6t,.) Thus then we see that when light passes (in this 

 theory) out of what is called a rarer medium into a 

 denser, or when the angle of refraction is less than that 

 of incidence, the velocity of propagation of the undulatory 

 movement is diminished, while on the corpuscular doctrine 

 it is increased, and vice versa. Thus, too, we see that on 

 the undulatory hypothesis the connexion betw&en refran- 

 gibility and velocity within the refracting medium is ini- 

 mediate and absolute, and consequently that it being cer- 

 tain, as we have shown, that light of all refrangibilities 

 travels equally fast in what 7ve <;<7//. empty space {i.e., 

 through the ether alone), it follows with equal certainty 

 that in material media the more refrangible rays are pro- 

 pagated slower than the less so ; and all, more slowly than 



