﻿Dispersion, 
  and 
  Surface- 
  Colour 
  of 
  Selenium. 
  (521 
  

  

  point 
  of 
  fact, 
  it 
  is 
  easily 
  seen 
  that 
  the 
  characteristic 
  plum- 
  

   colour 
  o£ 
  cyanine 
  films 
  is 
  due 
  rather 
  to 
  the 
  almost 
  complete 
  

   absence 
  of 
  reflexion 
  of 
  the 
  green 
  rays 
  than 
  to 
  strong 
  metallic 
  

   reflexion 
  of 
  any 
  particular 
  colour. 
  

  

  

  

  

  

  

  Fig. 
  

  

  5. 
  

  

  — 
  Curve 
  

  

  illustratii 
  

  

  g 
  

  

  Surface-colour 
  of 
  Cyanine. 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  ■3 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  "^ 
  

  

  

  

  

  

  •1 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  ^ 
  

  

  £*! 
  

  

  ># 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  A 
  

  

  r^ 
  

  

  

  

  ^ 
  

  

  s 
  

  

  L 
  / 
  

  

  

  

  

  

  

  

  

  •0 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  / 
  

  

  

  

  

  

  

  *J 
  

  

  «!>, 
  

  

  

  

  

  

  

  

  •8 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  / 
  

  

  / 
  

  

  

  

  

  

  

  

  > 
  

  

  fe 
  

  

  '+* 
  

  

  

  

  

  •7 
  

  

  

  ' 
  

  

  

  

  

  ?? 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  ^ 
  

  

  *t 
  

  

  

  

  

  

  

  

  

  

  <*• 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  <> 
  

  

  

  

  

  

  

  

  Wc 
  

  

  so 
  

  

  

  

  

  

  s 
  » 
  

  

  ^ 
  

  

  

  

  \? 
  

  

  '* 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  X 
  

  

  >r 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  ho, 
  

  

  ' 
  ' 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  >"© 
  

  

  

  

  

  

  

  

  4 
  

   3 
  

  

  •2 
  

   •1 
  

  

  

  

  

  

  

  

  

  

  

  \ 
  

  

  . 
  

  

  

  / 
  

  

  

  

  y 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  \ 
  

  

  

  V 
  

  

  > 
  

  

  

  ^ 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  \ 
  

  

  £ 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  ^ 
  

  

  

  

  

  1 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  400 
  

  

  In 
  the 
  case 
  of 
  selenium, 
  if 
  we 
  apply 
  the 
  reflexion 
  formula 
  

   to 
  the 
  data 
  that 
  have 
  been 
  obtained 
  thus 
  far, 
  we 
  obtain 
  a 
  

   curve 
  which 
  indicates 
  that 
  the 
  reflexion 
  increases 
  rapidly 
  

   with 
  decreasing 
  wave-length. 
  As 
  I 
  have 
  said 
  before, 
  multiple 
  

   reflexions 
  from 
  selenium 
  surfaces 
  give 
  no 
  trace 
  of 
  colour, 
  

   which 
  suggests 
  that 
  errors 
  may 
  exist 
  in 
  either 
  the 
  refraction 
  or 
  

   extinction 
  curve. 
  I 
  have 
  already 
  pointed 
  out 
  where 
  the 
  

   source 
  of 
  trouble 
  may 
  lie 
  in 
  the 
  case 
  of 
  the 
  latter 
  curve. 
  In 
  

   regard 
  to 
  the 
  refraction-curve, 
  it 
  may 
  be 
  considered 
  fairly 
  

   accurate 
  up 
  to 
  the 
  point 
  where 
  the 
  prism 
  determinations 
  

   stopped. 
  In 
  the 
  case 
  of 
  films 
  the 
  thickness 
  of 
  which 
  is 
  less 
  

   than 
  the 
  wave-length 
  of 
  light, 
  I 
  do 
  not 
  yet 
  feel 
  sure 
  that 
  the 
  

   displacement 
  of 
  interference-fringes 
  gives 
  us 
  a 
  true 
  measure 
  

   of 
  the 
  refractive 
  index. 
  I 
  hope 
  some 
  time 
  to 
  make 
  a 
  rigorous 
  

   test 
  of 
  this 
  point, 
  for 
  so 
  far 
  as 
  I 
  know 
  it 
  has 
  never 
  been 
  

   settled. 
  

  

  The 
  fact 
  that 
  the 
  dispersion-curve 
  of 
  cyanine 
  determined 
  

   by 
  the 
  interferometer 
  agreed 
  so 
  well 
  with 
  the 
  curve 
  obtained 
  

   by 
  means 
  of 
  a 
  thin 
  prism, 
  does 
  not 
  prove 
  conclusively 
  that 
  

   the 
  method 
  is 
  not 
  open 
  to 
  objection 
  ; 
  for 
  the 
  same 
  errors 
  

   might 
  occur 
  in 
  both 
  cases 
  : 
  in 
  other 
  words, 
  the 
  deviation 
  pro- 
  

   duced 
  by 
  a 
  prism, 
  the 
  base 
  of 
  which 
  was 
  only 
  a 
  wave-length 
  

   or 
  two 
  thick, 
  might 
  be 
  affected 
  by 
  phase 
  changes 
  at 
  the 
  

   surfaces. 
  

   1 
  Phil. 
  Mag. 
  S. 
  6. 
  Vol. 
  3. 
  No. 
  18. 
  June 
  1902, 
  2 
  T 
  

  

  