10 LECTURES TO SCIENCE TEACHERS. 



compound wave will be double the height of each of the 

 component waves. As one is gradually retarded or thrown 

 behind the other, the compound wave becomes less and less, 

 until at last, when the crest of one wave is opposite the hollow 

 of the other, the combined wave is absolutely destroyed. And 

 the same thing occurs if one wave is advanced in front of the 

 other. 



Now consider for a moment what has happened. There 

 we have two Nicol prisms. A ray of light after passing 

 through the first emerges polarised. Suppose for a moment 

 that the vibrations are horizontal; thk being so, if the 

 second prism is in a similar position, the light would pass 

 through. But if the second prism be turned round through 

 any angle it would only transmit the oblique vibrations, 

 partly horizontal and partly vertical, and when we turn it 

 round at right angles to the first, the light will be extinguished. 

 Supposing now that I introduce into this space a plate of 

 crystal, such as mica, upon it there falls a certain beam of 

 light whereof the vibrations are horizontal. On entering the 

 plate every ray of the beam will be divided into two, and the 

 vibrations in one of those rays will be in one direction, and 

 the vibrations in the other will be at right angles to the first. 

 Suppose, for convenience, the plate to be so placed that the 

 vibrations which it will transmit are at 45 to the horizontal, 

 say, north-east and south-east, and north-west and south-west, 

 or the dexter and sinister diagonal, as it is termed. Now when 

 the light emerges from this crystal we shall have two sets of 

 vibrations taking place in two rectangular directions, and 

 those forming one of those rays will be retarded behind the 

 other. The extent of the retardation is very small in abso- 

 lute amount, but, nevertheless, optically quite appreciable. 

 Supposing then, we cause these two sets of vibrations to fall 

 upon this second Nicol prism, or analyser ; the effect of that 

 is to resolve these two vibrations, or such parts of them as are 

 capable of being resolved, into a single plane. They may be 

 resolved into any direction at pleasure, by giving a suitable 

 direction to the analyser. Suppose that they be resolved into 

 a vertical plane, we shall have two sets of waves (portions of 

 the two oblique waves issuing from the plate) in a vertical 

 plane, one of which is somewhat behind the other. Now, as 

 Professor Stokes described to you yesterday, white light is 

 compounded of light of various kinds, i.e. of various wave- 



