C LECTURES TO SCIENCE TEACHERS. 



evidence for thinking that at different parts of the ray the 

 directions undergo abrupt changes. In polarised light, on 

 the contrary, whatever be the direction of vibration in one 

 part, the same is maintained throughout the whole ray. That 

 is the mechanical explanation of polarisation. If then the 

 transmission of a beam of light through a bundle of glass 

 plates such as I have described has the effect of bringing all 

 the vibrations into a certain direction dependent upon the 

 direction in which the plates face, it is not difficult to conclude 

 that the effect of a second bundle upon the polarisation of 

 the rays will be nil if it faces the same way as the first. To 

 fix our ideas, suppose that the vibrations of the rays reflected 

 from the first bundle be horizontal ; or in other words, that 

 its facing (or azimuth as it is termed) is such that it will 

 reflect only horizontal vibrations. Then it is clear that if 

 the second bundle have the same azimuth it will not affect the 

 direction or plane of polarisation, because it has received 

 those vibrations, and those only, which it is capable of re- 

 flecting. If, however, the second bundle be turned in the 

 manner described before, the direction of vibrations which it 

 is capable of reflecting will turn with it, and the vibrations 

 falling upon it will be oblique to those which it can reflect. 

 Now motion oblique to a given plane may be considered as 

 composed of two motions partly in the plane, and partly 

 perpendicular to it ; because if the oblique motion be replaced 

 by these two motions executed successively in suitable times, 

 the moving body will arrive at the same position in the same 

 time, whether it has passed by one of these paths or by the 

 other. The oblique motion is then said to have been resolved 

 in two directions, viz. parallel and perpendicular to the plane 

 in question. The second bundle of glass will therefore reflect 

 only that part of the vibrations falling on it which can be 

 resolved into its own plane ; the part resolved into the per- 

 pendicular direction will altogether fail to be reflected. The 

 amount of light reflected will consequently diminish according 

 as the second bundle has its azimuth turned away from that 

 of the first ; and when the turning has reached a right angle, 

 no light at all will be reflected. 



. Now for producing these effects we are not confined to 

 glass ; but we may use instead almost any other substance, 

 except metals ; and the only question is, at what angle inci- 

 dence must take place in order most effectually to polarise the 



