380 OPTICAL PROJECTION 



but this soon passes off, and the heat is applied till the 

 desired effect is produced. One crystal will last for many 

 demonstrations . 



Circularly polarised crystal figures are produced by placing 

 a quarter- wave plate in the ordinary stage. The black 

 brushes then disappear, nebulous lines taking their places ; 

 while the rings are dislocated half a wave in each alternate 

 quadrant. The rings may instead be analysed circularly, by 

 fitting a quarter-wave plate on either side of the focussing-lens, 

 K (fig. 190), with the same results. If the rings be both 

 polarised and analysed circularly, the brushes disappear 

 entirely, and as the analyser is rotated, the quadrants (or 

 halves in bi-axial rings) slide by each other, producing in the 

 two principal positions unbroken rings with no brushes or 

 interruption whatever. If at this point the quarter-wave be 

 rotated with the analyser, the unbroken character of the rings 

 is retained throughout all the rotation ; showing the perfectly 

 circular character of the polarisation. 



Spiral Figures were discovered by myself, 1 in a search 

 after phenomena which should more distinctly show the rela- 

 tion of bi-axial to uni-axial crystals, and of the two axes of a bi- 

 axial to the prismatic axis or, in short, that the axis of a uni- 

 axial was simply a case of the coincidence of two axes. For 

 obvious reasons this was most likely to be brought about by the 

 two circular waves concerned in rotary polarisation ; and it 

 seemed worth while to seek for such demonstration, since 

 when polarised and analysed circularly, one single axis of a 

 bi-axial gives as unbroken a circle as a uni-axial. I sought 

 for phenomena which might show that each axis of a bi-axial 

 was only one sex, as it were, of a combination, both of which 

 were found in a uni-axial. This is shown by placing in the 

 ordinary stage of the polariscope a quartz plate 7^ mm. thick, 

 and introducing between the crystal and the analyser a 



1 See Proceedings of the Physical Society for November 12, 1881, or Phil, 

 Mag. January 1882, 



