1889.J on Optical Torque. 477 



Now, in the case of waves of natural light from all ordinary 

 sources— sun, stars, candles, gas-flames, or electric lights — the waves 

 emitted are not found to be polarised. That is to say, their motions 

 are not executed in any particular plane, nor even in any particular path 

 of any kind ; they appear to be absolutely heterogeneous at least so far 

 as this, that no vibration of the millions of millions emitted in a second 

 of time is followed by more (on the average) than about 50,000 vibrations 

 of a similar sort, executed along a similar j^ath — the plane of the polari- 

 sation, if any, changing after the lapse of such an incredibly short 

 time that for most purposes the vibrations in ditferent directions are as 

 inextricably mixed as if they had all been simultaneously jumbled up. 

 Since, then, natural light is non-polarised or miscellaneous, the pro- 

 duction of polarised light must be brought about by the employment of 

 polarising apparatus or agents which will so operate on or affect the 

 mixed waves as to bring their vibrations into one direction — or, what 

 amounts to the same thing, transmit the light whilst destroying or 

 absorbing those parts of the vibrations which are executed across the 

 desired line of vibration. So we have jjolarisers consisting of tour- 

 maline slices ; oblique bundles of thin glass plates ; black - glass 

 reflectors ; and Nicol prisms cut from calc-spar. About the two 

 latter I may be permitted a passing word presently. These objects 

 polarise, i.e. tiu-n into one plane, the vibrations of light falling upon 

 them. A rough mechanical illustration may here be permitted me. 

 A long indiarubber cord is passed through the open ends of a box 

 provided with vertical partitions. Fig. 1 shows the arrangement. 

 These partitions confine the motion of the cord, and effectually polarise 

 the vibrations which I now impart to the cord by shaking the end of 

 it to and fro. If the partitions are vertical, the box polarises, into 

 vertical vibrations only, the miscellaneous vibrations which are sent 

 to it. If rotated until its partitions ai-e horizontal, it polarises the 

 vibrations into a horizontal position. 



Let us now turn to the optical analogue of this experiment. The 

 large Nicol prism which I introduce into the field of the electric- 

 light lantern, polarises the light, so that the vibrations are executed 

 dimply in an up-and-down direction. Your eye will not detect this, 

 he motion being millions of times too rapid. To detect the direction 

 m analyser is necessary. For this purpose a second apparatus of 

 he same sort is used, for then, by crossing the positions of the two, 

 ho whole of the light is cut off"; the second Nicol prism, if set so as 

 o transmit only horizontal vibrations, cutting off the vertical 

 dbrations that are sent through the first prism. So, while the first 

 )rism serves as a polariser, the second serves as an analyser to detect 

 )y cutting them off when turned to the proper position, the direction 

 •f the polarisation which had been previously impressed by the first 

 -rism. 



Here I may illustrate the action of the analyser for determininc^ 

 he plane of polarisation of the vibrations by the extinction which it 

 reduces when turned to the crossed position. For this i^urpose I 



2 K 2 



