﻿368 
  Messrs. 
  F. 
  Bryson 
  and 
  J. 
  Logie 
  on 
  the 
  

  

  place 
  was 
  a 
  concave 
  grating 
  of 
  10 
  ft. 
  radius 
  of 
  curvature, 
  

   ruled 
  by 
  the 
  late 
  Lord 
  Blythswood. 
  It 
  was 
  hoped 
  that 
  the 
  

   grating 
  would 
  prove 
  powerful 
  enough 
  to 
  analyse 
  the 
  line 
  

   and 
  that 
  the 
  results 
  would 
  be 
  of 
  the 
  more 
  interest, 
  as 
  most 
  

   of 
  the 
  previous 
  work 
  has 
  been 
  done 
  with 
  echelon 
  spectro- 
  

   scopes 
  or 
  Lunimer 
  plates. 
  Also 
  the 
  simplicity 
  of 
  the 
  

   interpretation 
  of 
  results 
  which 
  a 
  grating 
  secures 
  made 
  its 
  

   use 
  desirable. 
  The 
  grating 
  was 
  mounted 
  in 
  a 
  manner 
  ad- 
  

   vocated 
  by 
  Eagle 
  * 
  as 
  being 
  specially 
  suitable 
  for 
  work 
  of 
  this 
  

   nature, 
  giving 
  great 
  steadiness, 
  and 
  constancy 
  of 
  temperature. 
  

   Unfortunately 
  it 
  was 
  found 
  that 
  the 
  apparatus 
  had 
  not 
  

   sufficient 
  resolving 
  power 
  to 
  give 
  clear 
  and 
  reliable 
  results. 
  

   The 
  ruled 
  surface 
  of 
  the 
  grating 
  is 
  1\ 
  ins. 
  by 
  2 
  ins., 
  which 
  

   is 
  much 
  smaller 
  than 
  that 
  of 
  the 
  grating 
  used 
  by 
  Gale 
  and 
  

   Lemon 
  | 
  in 
  their 
  work 
  on 
  the 
  analysis 
  of 
  the 
  mercury 
  lines. 
  

  

  An 
  echelon 
  spectroscope 
  of 
  26 
  plates, 
  each 
  9*73 
  mm. 
  thick, 
  

   was 
  then 
  employed. 
  The 
  constants 
  of 
  this 
  echelon 
  are 
  

   given 
  in 
  a 
  paper 
  contributed 
  to 
  the 
  Royal 
  Society 
  by 
  

   Professor 
  Gray 
  and 
  Dr. 
  Stewart 
  J. 
  The 
  green 
  line 
  in 
  the 
  

   spectrum 
  was 
  isolated 
  in 
  certain 
  cases 
  by 
  using 
  a 
  Bilger 
  

   constant 
  deviation 
  spectroscope, 
  and 
  at 
  other 
  times 
  by 
  means 
  

   of 
  a 
  large 
  prism 
  placed 
  between 
  the 
  echelon 
  and 
  the 
  telescope 
  

   or 
  camera. 
  

  

  Sources. 
  — 
  Three 
  sources 
  of 
  light 
  were 
  employed: 
  — 
  (1) 
  A 
  

   barometric 
  U-tube, 
  of 
  Jena 
  glass, 
  with 
  the 
  arc 
  across 
  the 
  

   vacuum. 
  This 
  had 
  the 
  advantage 
  that 
  the 
  vapour 
  tension 
  

   could 
  be 
  easily 
  determined 
  from 
  the 
  height 
  of 
  the 
  mercury 
  

   column, 
  and 
  the 
  length 
  of 
  arc 
  could 
  be 
  increased 
  or 
  

   diminished 
  by 
  raising 
  or 
  lowering 
  the 
  reservoir 
  connected 
  

   to 
  one 
  end 
  of 
  the 
  U-tube. 
  Into 
  the 
  other 
  closed 
  end 
  was 
  

   sealed 
  a 
  platinum 
  electrode. 
  We 
  found 
  that 
  after 
  a 
  few 
  

   weeks' 
  use 
  the 
  variations 
  of 
  temperature 
  caused 
  the 
  glass 
  

   to 
  crack 
  at 
  the 
  point 
  where 
  the 
  electrode 
  was 
  sealed 
  through, 
  

   causing 
  a 
  leak 
  and 
  thus 
  spoiling 
  the 
  vacuum. 
  This 
  form 
  of 
  

   lamp 
  was 
  ultimately 
  discarded. 
  (2) 
  A 
  small 
  exhausted 
  arc 
  

   lamp, 
  with 
  an 
  automatic 
  starter, 
  devised 
  by 
  Anderson 
  and 
  

   Burnside, 
  and 
  described 
  in 
  a 
  communication 
  recently 
  made 
  

   to 
  the 
  Royal 
  Society 
  of 
  Edin 
  burgh 
  §. 
  This 
  lamp 
  was 
  used 
  

   fairly 
  extensively 
  and 
  gave 
  very 
  good 
  results. 
  (3) 
  A 
  Com- 
  

   mercial 
  Cooper 
  Hewitt 
  27- 
  inch 
  lamp. 
  This, 
  though 
  of 
  

   greater 
  candle-power, 
  was 
  of 
  no 
  higher 
  intrinsic 
  brilliance 
  

   than 
  the 
  small 
  lamp. 
  

  

  * 
  Astrophys. 
  Journal; 
  xxxi. 
  p. 
  120 
  (1910). 
  

  

  t 
  Ibid. 
  xxxi. 
  p. 
  78. 
  

  

  t 
  Proc. 
  Roy. 
  Soc. 
  lxxii. 
  No. 
  477, 
  p. 
  16 
  (1903). 
  

  

  § 
  Proc. 
  Roy. 
  Soc. 
  Edin. 
  vol. 
  xxxiii. 
  pt. 
  1, 
  p. 
  117 
  (1913). 
  

  

  