666 PROCEEDINGS OF THE 



in the crystal section is the direction of vibration of light with ' 

 the greater velocity, and which that of light with the less velocity, 

 and determine, at the same time, the relative retardation, or, in 

 other words, how far the slower moving vibrations have lagged 

 behind the faster. (This is usually measured in micro-milli- 

 metres, or millionths of a millimetre.) For the purpose of 

 making these determinations we insert both nicols in the cross 

 position and rotate the stage till the direction of vibration is 

 diagonal to those of the nicols. The two vibrations which pass 

 the lower nicol are now resolved along the two directions of 

 vibration of the section. If there were no relative retardation 

 they would in emergence recombine to form a vibration parallel 

 to the same direction as before, and would be extinguished by 

 the upper nicol. As a result of the relative retardation, however, 

 the various colours of the spectrum are transmitted in different 

 degrees, so that the compound tints known as interference 

 colours are obtained. These are dependent upon the amount 

 of it, which is usually about the same for all colours of the 

 spectrum. 



Interference colours commence with complete darkness at 

 zero relative retardation and pass through grey, white, yellow, 

 orange and red, which last is seen when the relative retardation 

 reaches 550 micro-m.m. These constitute the colours of the first 

 order. Then follow r purple, violet, blue, green, yellow and red 

 up to a relative retardation of 1,000. These are the second 

 order. Every addition of 550 micro-m.m. corresponds to another 

 order with a similar succession of colours, gradually becoming 

 more complex until they are delicate shades of green and pink, 

 and with a relative retardation of about 4,000 micro-m.m. they 

 slowly pass into white light. 



If one nicol be rotated through a quarter turn so that the 

 directions of vibrations of the two nicols are parallel, the colours 

 are seen to pass through brown, red and blue to the yellowish 

 green, marking the end of the first order at 550. Then the 

 second, and gradually the colours fade into white light, exactly 

 as with crossed nicols. 



The amount of relative retardation in a crystal section may 

 be roughly estimated directly from the interference colours 

 between crossed and parallel nicols by comparison with tables 

 or lithographic plates of colours giving the corresponding 



