DOUBLE REFRACTION. 117 



If a card be passed along tlie under surface of the crystal, in the direction 

 R'R, it will cut oft' the ray OP before interfering with OQ. The image O', 

 which is most distant from the card, is therefore first to disappear — a phenome- 

 non very striking when seen fir the first time. The card employed in this 

 experiment shouUl be dull black in order to produce the best effect, otherwise it 

 is too conspicuous itself. 



Wiien a ray of light, after having passed through one crystal and having been 

 divid(>d into two distinct emergent rays, is allowed to fall upon another similar 

 and equal crystal similarly situated, the effect, as might naturally be expected, 

 will be to increase the separation of the rays to the same extent as Avould have 

 occurred had both the crystals been united in one. But if the second crystal 

 be turned around the direction of the ray as an axis, other phenomena make 

 their appearance, the character of which depends on the amount of turning. In 

 speaking of this kind of revolution it will be convenient to employ the term 

 azimutli. By this word is meant direction in space in a plane at right angles 

 to any axial line. The term is adopted from astronomy and geodesy, in which 

 sciences the assumed axial line is the vertical, and the azimuthal plane the 

 horizon. 



In -the case in hand, if we completely reverse the position of the second 

 crystal in azimuth, that is to say, turn it round ISO*^, it will reverse the refract- 

 ing effect of the first crystal and reunite the two rays, which Avill emerge as 

 one. If we turn it only 90° in azimuth the separation of the rays will con- 

 tmue, but that which was the extraordinary ray in the first crystal will become 

 the ordinary in the second, and vice versa. Accordingly, if the original inci- 

 dence is perpendicular, the ray which follows the normal in the first crystal will 

 be bent at the surface of the second, and that which is bent at the surface of the 

 first will follow the normal on entering the second. 



At any azimuth differing from the original position more or less than 90° or 

 180° there will be seen four emergent rays, of which two will usually posse?* 

 a greater intensity than the other two. When the change of position of the 

 second crystal is but slight, the two original rays will be vivid; but, in a line at 

 right angles to that which connects them, two very faint ones will appear, nearer 

 together than the original two. As the rotation advances these new rays will 

 gain in strength, while the other two grow less intense. At the azimuth of 45° 

 the four will be equal and equidistant. Beyond 45° the original rays go on 

 fading and the new ones increasing in brightness, until, at 90°, the former become 

 entirely extinct and the new ones remain alone. Beyond 90° again another 

 faint pair appear, Avhich go on, as before, increasing in brightness, at the expense 

 of the companion pair, up to the azimuth 135°, when the four are again equal. 

 Beyond 135° this second new pair still continue to gain strength and to approach 

 each other, till, at the azimuth 180°, they reunite into one, and the others in their 

 turn vanish. In the figure following, these successive phases are shown as 

 they appear upon a screen when the experiment is performed in a dark room. 

 They are circumscribed by the Outlines of the two rhombs in their relative suc- 

 cessive positions. 



The phenomena of double refraction were carefully studied by the celebrated 

 IluygheUs, who devised a physical theory for their explanation, which has been 

 pronounced by Brewster to be one of the most splendid of the triumphs of genius 

 Avhich illustrate the history of science. His theory did not, however, exlend to 

 the explanation of the remarkable appearances last described, a\ Inch present 



