﻿and 
  Diffraction 
  Telescopes. 
  521 
  

  

  exposure 
  and 
  development 
  was 
  the 
  same, 
  and 
  both 
  were 
  taken 
  

   on 
  the 
  same 
  plate. 
  Reproductions 
  of 
  the 
  two 
  images 
  are 
  

   shown 
  in 
  fig. 
  3. 
  It 
  will 
  be 
  noted 
  that 
  the 
  brilliant 
  one 
  

  

  Fiar. 
  3. 
  

  

  (or 
  phase-reversal 
  image) 
  is 
  larger 
  than 
  the 
  other. 
  This 
  is 
  

   due 
  to 
  the 
  increase 
  in 
  the 
  focal 
  leugth 
  to 
  w 
  T 
  hich 
  I 
  have 
  

   alluded, 
  and, 
  other 
  things 
  being 
  equal, 
  would 
  make 
  the 
  

   illumination 
  less. 
  The 
  image 
  obtained 
  from 
  the 
  outer 
  surface 
  

   of 
  the 
  plate 
  is 
  so 
  faint 
  that 
  it 
  is 
  hardly 
  perceptible. 
  

  

  I 
  have 
  made 
  a 
  very 
  cursory 
  examination 
  of 
  these 
  phase- 
  

   reversal 
  reflecting 
  plates 
  with 
  polarized 
  light, 
  and 
  find 
  that 
  if 
  

   the 
  light 
  falling 
  on 
  the 
  prism 
  be 
  plane-polarized, 
  rotation 
  of 
  

   the 
  nicol 
  causes 
  fluctuations 
  in 
  the 
  brilliancy 
  of 
  the 
  image. 
  

   The 
  effect, 
  however, 
  varies 
  with 
  the 
  angle 
  of 
  incidence 
  ; 
  at 
  an 
  

   angle 
  of 
  40° 
  with 
  the 
  normal 
  of 
  the 
  reflecting 
  surface 
  no 
  effect 
  

   is 
  produced 
  by 
  rotating 
  the 
  nicol. 
  When 
  the 
  angle 
  is 
  greater 
  

   than 
  this, 
  partial 
  extinction 
  of 
  the 
  image 
  is 
  produced 
  when 
  

   the 
  plane 
  of 
  polarization 
  is 
  perpendicular 
  to 
  the 
  plane 
  of 
  

   reflexion 
  ; 
  when 
  the 
  angle 
  is 
  less 
  than 
  40°, 
  the 
  reverse 
  is 
  the 
  

  

  case. 
  The 
  field 
  about 
  the 
  image 
  grows 
  bright 
  when 
  the 
  

  

  © 
  © 
  © 
  

  

  central 
  image 
  weakens, 
  and 
  vice 
  versa. 
  This 
  requires 
  a 
  much 
  

   more 
  complete 
  investigation 
  than 
  I 
  have 
  had 
  time 
  to 
  give 
  it 
  

   yet, 
  but 
  is 
  doubtless 
  caused 
  by 
  the 
  variations 
  in 
  the 
  phase- 
  

   change 
  due 
  to 
  variations 
  in 
  the 
  angle 
  of 
  incidence 
  which 
  

   Quincke 
  observed. 
  

  

  Lord 
  Rayleigh 
  has 
  drawn 
  attention 
  to 
  the 
  fact 
  that 
  if 
  it 
  

   were 
  possible 
  to 
  construct 
  a 
  lamina-grating 
  in 
  which 
  an 
  arbi- 
  

   trary 
  retardation 
  could 
  be 
  introduced 
  at 
  every 
  part 
  of 
  the 
  

   aperture 
  all 
  the 
  light 
  could 
  be 
  concentrated 
  in 
  an}^ 
  desired 
  

   spectrum. 
  This 
  might 
  be 
  accomplished, 
  as 
  Prof. 
  Crew 
  has 
  

   suggested 
  to 
  me, 
  by 
  shading 
  the 
  original 
  drawing 
  : 
  possibly 
  

   photographing 
  a 
  wire-grating 
  on 
  a 
  white 
  ground, 
  illuminated 
  

  

  