350 



POLARISATION. 



On these suppositions a transverse section of the cylinder 

 gives the simplest combination, since the effects of the super- 

 posed elements become added together 

 as in a simple crystal ; moreover, the 

 positions of the ellipses of elasticity 

 effective upon a radius agree with each 

 other (Fig. 200). The combined effect 

 is at once evident from these facts. 

 Each diameter of the section acts as a 

 crystal needle (or, in thicker sections, as 

 a plate of crystal placed vertically), and 

 the interference colours, which come 

 into view successively on rotating the 

 crystal needle round a perpendicular 



axis, appear side by side at the same time upon the sectional sur- 

 face. Two diametral zones a b and c d, in which the axes of the 

 ellipses of elasticity lie in the polarising planes P P and N N of 

 the Nicols, consequently act as single-refracting media ; without 

 the selenite plate they appear black, and with it they appear 

 illuminated by its normal interference colour. The middle lines 

 of the intermediate quadrants of necessity, therefore, show the 

 brightest colours, and, in combination with a selenite plate, pairs 

 of acceleration or retardation colours, according as the homologous 

 axes of the ellipses coincide or intersect at right angles. Upon a 

 selenite plate, red I., whose ellipse of elasticity takes the position 

 represented in Fig. 201, the quadrants of 

 the cylinder denoted by A would therefore 

 increase the colour, whilst those denoted by 

 S would diminish it. 



Hence, in transverse sections of cylinders 

 and hollow cylinders, the axial position of 

 the effective ellipses of elasticity is easily 

 discovered. We will only add, that all 

 hitherto known cases so far agree, that the 

 one axis is situated radially, the other at a 

 tangent, while the neutral zones which form the dark cross always 

 lie in the polarising planes of the Nicols. The deviation (from 

 this) shown in Figs. 200 and 201 is not therefore actually 

 observed, but is merely assumed for the sake of generality. 



FIG. 201. 



