J. tr. HAEFOED POLAEISATION OF LIGHT. 153 



"We will now proceed to exemplify and explain in a more precise 

 and practical way these general remarks. 1st. To investigate the 

 effect of reflexion. A beam of ordinary light let fall on a reflecting 

 surface, not metallic — say glass — at an angle of 56j° of incidence, 

 or making an angle with the glass of 33|'^, is partly reflected and 

 partly transmitted (Fig. 3), as, in fact, would be the case at any 

 other angle of reflexion. But at this particular angle the light 

 that is reflected becomes completely polarised. I say "completely" 

 because at any other angle only partial polarisation takes place ; 

 and the angle of 56^'^ is therefore called the polarising angle for 

 glass,* other reflecting bodies having each a particular angle of 

 maximum polarisation. Now this light so reflected is capable of 

 being reflected again from a second reflecting plate at any angle of 

 position, so that the reflexion takes place in the same plane, but will 

 not bear reflexion in a plane de-viating from the plane of incidence 

 and fij'st reflexion, and its inability of reflexion increases up to the 

 angle of 90 degrees, a right angle with the first plane, at which 

 angle no light is reflected by the second reflector. The light re- 

 flected by the first plate at the given angle is therefore said to be 

 polarised in the plane of reflexion. Now what becomes of the 

 portion of light that passes through the first glass plate ? By 

 testing it in a similar way by a second reflector, it is found to be 

 capable of second reflexion just in the directions in which it refused 

 to pass by reflexion from the first reflecting surface, and incapable 

 of reflexion in the original plane of incidence. Thus, then, the 

 light that passes through the first plate is said to be polarised by 

 refraction. It is necessary to state here that refraction through a 

 single plate is insufficient to polarise the whole of the transmitted 

 light; but by the use of several plates superposed, called "a 

 bundle," the efiiect is proportionately increased until nearly entire 

 polarisation ensues. We have now to notice a further and most 

 interesting form of polarisation attendant on what is known as 

 double refraction. The most remarkable substance exemplifying 

 this property is Iceland spar. A beam of light in passing through 

 this mineral in any direction (except one, hereafter to be considered) 

 is divided into two (Fig. 4). An object, therefore, seen through a 

 plate of this crystal, appears doubled. A small hole in a card, or a 

 black spot on paper, will appear doubled ; and what concerns our 

 subject is this — that the two beams into which the original one is 

 divided are polarised at right angles to each other. What glass 

 plates and bundles have efl'ected by reflexion and transmission re- 

 spectively, is in the case of Iceland spar effected by transmission 

 solely, through the separation of the beam of light into two. 



This subject of double refraction opens up a wide field of re- 

 search in connexion with Mineralogy and Crystallography, and 

 also has to do with a great class of most interesting and beautiful 



* The polarising angle varies, however, according to the index of refraction of 

 the glass. Deschanel, in his ' Elementary Treatise on Natural Philosophy ' 

 (edition by Prof. Everett, page 1034), states that the polarising angle for crown- 

 glass varies from 56° 51' to 57'' 23', and for flint-glass from 57'' 3(3' to 58" 40'.— Ed. 



VOL. I. — FT. VI. 12 



