1897.] on the Polarisation of the Electric Bay. 305 



duced by unannealed glass may be sbown by this piece of solid 

 paraffin, wbicb was cast in a mould, and chilled unequally by a freez- 

 ing mixture. One of these blocks was cast two years ago, and it has 

 still retained its unannealed property. This effect may even be 

 shown without any special preparation. Pieces of glass or ebonite, 

 too, are often found sufficiently strained to exhibit double refraction. 



Phenomena of Double Absorption. 



Being desirous of making a crystal polariser, I naturally turned 

 *to tourmaline, but was disappointed to find it utterly unsuitable as 

 a polariser. There is a difference in transparency in directions 

 parallel and perpendicular to the length, but even a considerable 

 thickness of the crystal does not completely absorb one of the two 

 rays. Because visible light is polarised by absorption by tourma- 

 line, it does not follow that all kinds of radiation would be so 

 polarised. The failure of tourmaline to polarise the Rontgen rays 

 is therefore not unexpected, supposing such rays to be capable of 

 polarisation. 



It was a long time before I could discover crystals which acted 

 as electric tourmalines. In the meanwhile I found many natural 

 substances which produced polarisation by selective unilateral ab- 

 sorption. For example, I found locks of human hair to polarise the 

 electric ray. I have here two bundles of hair ; I interpose one 

 at 45°, and you observe the depolarisation effect. The darker 

 specimen seems to be the more efficient. Turning to other substances 

 more easily accessible, I found vegetable fibres to be good polarisers. 

 Among these may be mentioned the fibres of aloes (^Agave), rhea 

 (Boehmeria nivea), pine-apple [Ananas sativus), plantain (3Iusa para- 

 disiaca). Common jute [Corchorus capsularis) exhibits the property 

 of polarisation in a very marked degree. I cut fibres of this material 

 about 3 cm. in length, and built with them a cell with all the fibres 

 parallel. I subjected this cell to a strong pressure under a press. 



1 thus obtained a compact cell 3 cm. by 3 cm. in area, and 5 cm. in 

 thickness. This was mounted in a metallic case, with two openings 



2 cm. by 2 cm. on opposite sides for the passage of radiation. This 

 cell absorbs vibrations parallel to the length of the fibres, and trans- 

 mits those perpendicular to the length. Two such cells could thus 

 be used, one as a polariser and the other as an analyser. 



Turning to crystals, I found a large number of them exhibiting 

 selective absorption in one direction. Of these nemalite and cryso- 

 tile exhibit this property to a remarkable extent. Nemalite is a 

 fibrous variety of brucite ; crysotile being a variety of serpentine. 

 The direction of absorption in these cases is parallel to the length, 

 the direction of transmission being perpendicular to the length. I 

 have here a piece of crysotile, only one inch in thickness. I adjust 

 the polariser and the analyser parallel, and interpose the crysotile 

 with its length parallel to the electric vibration. You observe that 



