160 ELEMENTS OF LABORATORY WORK 



light this plane must, from our observations, be perpendicular 

 to the direction of propagation. Confirmation of this may be 

 obtained from the polarisation of light. 



In order to measure the wave-length X of a given kind of 

 light it is necessary to know the distance x between two con- 

 secutive bright bands, the distance a between the slit and the 

 eye-piece, and the distance C between the two virtual images 

 formed by the two halves of the prism. Then, 



By varying the light, or interposing coloured glasses 

 between the slit and the bi-prism, the variation in wave-length 

 may be measured by the same formula, the quantity x being 

 the only variable. We shall find that the colours in the order 

 in which they are presented in the spectrum vary in wave- 

 length, the smallest wave-lengths being at the violet end, and 

 the longest at the red end, the intermediate colours having 

 intermediate wave-lengths. 



102. Explanation of Rectilinear Propagation. Although 

 the wave theory of light has simply and reasonably explained 

 several phenomena, we have got to find some explanation for the 

 observed rectilinear propagation of light, and for the observed 

 laws of reflexion and refraction. It is contrary to observation 

 that a source of light should travel in one direction only. It is 

 visible in all directions except so far as opaque matter pre- 

 vents its passage. It would also be inconsistent with our 

 assumption of the existence of a homogeneous ether to suppose 

 that a disturbance, set up by a radiating body, does not travel 

 in the same manner, and with the same speed, in all direc- 

 tions in the same isotropic medium. We have, in fact, since 

 the sphere is the only body which is symmetrical in all direc- 

 tions with regard to space, a series of spherical waves extend- 

 ing further and further until obstructed or modified. When 

 the surface of another medium is reached, we have in general 

 two new waves set up, each different in direction from the 

 original. One is the reflected wave, the other the refracted 

 wave. The relation in the amplitude of the vibrations, by 



