﻿90 
  Ewell 
  — 
  Rotatory 
  Polarization 
  of 
  Light 
  

  

  in 
  

  

  I 
  have 
  found 
  but 
  one 
  other 
  uiemoir 
  directly 
  on 
  this 
  subject. 
  

   In 
  a 
  paper 
  on 
  torsion 
  Wertheim^ 
  describes 
  the 
  optical 
  charac- 
  

   teristics 
  of 
  twisted 
  glass 
  cylinders. 
  Glass 
  rods, 
  about 
  30^°* 
  

   long 
  and 
  2*^"" 
  square 
  were 
  annealed 
  until 
  they 
  showed 
  only 
  a 
  

   faint 
  cross 
  when 
  viewed 
  between 
  crossed 
  IS'icols. 
  A 
  double 
  

   refracting 
  plate 
  giving 
  the 
  sensitive 
  tint 
  was 
  placed 
  between 
  

   the 
  polarizer 
  and 
  the 
  end 
  of 
  the 
  glass 
  bar. 
  If 
  the 
  analyzer 
  

   was 
  set 
  at 
  right 
  angles 
  to 
  the 
  polarizer, 
  of 
  the 
  four 
  quadrants 
  

   into 
  which 
  the 
  dark 
  cross 
  divided 
  the 
  cross 
  section 
  of 
  the 
  

   prism, 
  two 
  diagonally 
  opposite 
  were 
  green 
  and 
  the 
  other 
  two 
  

   yellow. 
  If 
  the 
  prism 
  was 
  twisted, 
  the 
  cross 
  left 
  the 
  center, 
  

   which 
  became 
  yellow 
  if 
  the 
  twist 
  was 
  to 
  the 
  right 
  and 
  green 
  if 
  

   to 
  the 
  left. 
  However, 
  almost 
  identical 
  results 
  were 
  obtained 
  

   when 
  the 
  bar 
  was 
  simply 
  rotated 
  to 
  the 
  right 
  or 
  left 
  without 
  

   twist, 
  whence 
  Wertheim 
  concluded 
  that 
  torsion 
  simply 
  pro- 
  

   duced 
  double 
  refraction, 
  positive 
  or 
  negative, 
  according 
  to 
  the 
  

   direction 
  of 
  the 
  twist. 
  Wertheim 
  was 
  much 
  disappointed 
  in 
  

   his 
  experiments 
  and 
  considered 
  that 
  they 
  decided 
  nothing 
  about 
  

   the 
  properties 
  of 
  perfectly 
  homogeneous 
  bodies 
  subjected 
  to 
  

   torsion 
  and 
  that 
  they 
  neither 
  verified 
  nor 
  disproved 
  Neumann's 
  

   mathematical 
  results. 
  

  

  An 
  interesting 
  parallelism 
  with 
  the 
  action 
  of 
  light 
  in 
  a 
  

   twisted 
  medium 
  is 
  Bose'sf 
  experiment 
  in 
  which 
  he 
  found 
  rota- 
  

   tion 
  of 
  the 
  plane 
  of 
  polarization 
  of 
  electro-magnetic 
  waves, 
  

   traversing, 
  a 
  twisted 
  cylinder 
  of 
  jute 
  fibres 
  parallel 
  to 
  the 
  axis. 
  

   If 
  the 
  twist 
  of 
  the 
  bundle 
  of 
  jute 
  fibres 
  was 
  reversed 
  the 
  plane 
  

   of 
  polarization 
  was 
  rotated 
  in 
  the 
  opposite 
  direction, 
  but 
  Bose 
  

   does 
  not 
  state 
  whether 
  the 
  rotation 
  of 
  the 
  plane 
  of 
  polarization 
  

   was 
  in 
  the 
  same 
  direction 
  as 
  the 
  twist 
  or 
  in 
  the 
  opposite 
  direc- 
  

   tion. 
  The 
  analogy 
  to 
  light 
  in 
  a 
  twisted 
  medium 
  must 
  not 
  be 
  

   carried 
  too 
  far 
  on 
  account 
  of 
  the 
  great 
  difference 
  in 
  wave 
  

   length 
  of 
  the 
  disturbances 
  considered 
  and 
  in 
  the 
  character 
  of 
  

   the 
  media. 
  

  

  I 
  first 
  studied 
  the 
  effect 
  of 
  torsion 
  on 
  the 
  optical 
  properties 
  

   of 
  a 
  rectangular 
  glass 
  bar 
  made 
  by 
  Duboscq 
  of 
  Paris, 
  15^"" 
  

   by 
  12"'"" 
  by 
  7°"". 
  The 
  ends 
  of 
  the 
  bar 
  were 
  placed 
  in 
  thin 
  

   wooden 
  cylinders 
  with 
  rectangular 
  apertures 
  slightly 
  larger 
  

   than 
  the 
  bar. 
  These 
  cylinders 
  were 
  grasped 
  by 
  iron 
  clamps 
  

   attached 
  to 
  long 
  levers 
  by 
  which 
  the 
  torsion 
  was 
  applied. 
  The 
  

   clamps 
  could 
  grasp 
  the 
  cylinders 
  in 
  any 
  position 
  and 
  thus 
  the 
  

   bar 
  could 
  be 
  studied 
  in 
  different 
  azimuths. 
  This 
  bar 
  showed 
  

   no 
  dai'k 
  cross 
  when 
  placed 
  between 
  crossed 
  analyzer 
  and 
  polar- 
  

   izer, 
  but 
  was 
  dark 
  along 
  the 
  edges. 
  The 
  bar 
  was 
  tested 
  in 
  

   both 
  the 
  Laurent 
  and 
  Scheibler 
  polarimeters, 
  and 
  although 
  the 
  

   phenomena 
  were 
  not 
  as 
  distinct 
  as 
  was 
  desirable, 
  and 
  varied 
  in 
  

  

  ^ 
  Ann. 
  de 
  cliim. 
  et 
  de 
  pliys., 
  iii, 
  1. 
  1857. 
  

  

  f 
  Proceedings 
  of 
  the 
  Royal 
  Society, 
  vol. 
  Ixiii, 
  1898. 
  

  

  