﻿958 
  Dr. 
  Gr. 
  Johnstone 
  Stoney 
  on 
  

  

  Lacus 
  is 
  seen 
  as 
  a 
  dark 
  object 
  with 
  brighter 
  surroundings 
  ; 
  

   but 
  this 
  difference 
  is 
  immaterial, 
  since 
  the 
  optical 
  conditions 
  

   for 
  seeing 
  a 
  bright 
  object 
  upon 
  a 
  dark 
  background, 
  and 
  a 
  

   dark 
  object 
  of 
  the 
  same 
  shape 
  and 
  size 
  upon 
  a 
  bright 
  back- 
  

   ground, 
  are 
  known 
  to 
  be 
  the 
  same. 
  

  

  67. 
  The 
  object 
  on 
  the 
  planet 
  which 
  we 
  have 
  chosen 
  for 
  our 
  

   first 
  experiment 
  being 
  a 
  round 
  object 
  of 
  such 
  a 
  size 
  that 
  it 
  

   subtends 
  angle 
  e 
  (2 
  ,,# 
  062648) 
  at 
  the 
  distance 
  of 
  the 
  earth 
  will 
  

   be 
  adequately 
  represented 
  iii 
  the 
  OEA 
  (observatory 
  experi- 
  

   mental 
  apparatus) 
  at 
  :', 
  by 
  a 
  round 
  hole 
  I 
  mm. 
  in 
  diameter, 
  

   in 
  the 
  screen 
  of 
  copper-foil 
  Z', 
  when 
  the 
  carpenter's 
  square 
  

   which 
  support- 
  the 
  screen 
  has 
  been 
  pushed 
  into 
  its 
  place 
  

   near 
  the 
  right-hand 
  surface 
  of 
  lens 
  [/, 
  with 
  z' 
  opposite 
  to 
  the 
  

   middle 
  u\' 
  the 
  lens. 
  This 
  is 
  to 
  be 
  illuminated 
  by 
  the 
  single 
  

  

  stellade 
  of 
  light 
  issuing 
  from 
  pro-star 
  /, 
  and 
  the 
  observer 
  can 
  

   bring 
  the 
  direction 
  in 
  which 
  this 
  stellade 
  is 
  incident 
  upon 
  

   object 
  z' 
  into 
  any 
  of 
  the 
  directions 
  along 
  whichlight 
  from 
  the 
  

   various 
  parts 
  of 
  a 
  source 
  of 
  diffused 
  Light, 
  such 
  as 
  <r, 
  would 
  

   reach 
  that 
  object. 
  When 
  the 
  incident 
  lio-ht 
  reaches 
  screen 
  r 
  /J, 
  

   most 
  of 
  the 
  light 
  is 
  -topped 
  by 
  that 
  screen, 
  the 
  only 
  light 
  

   which 
  is 
  allowed 
  to 
  advance 
  beyond 
  the 
  screeu 
  being 
  that 
  

   which 
  passes 
  through 
  the 
  found 
  opening 
  ~'. 
  We 
  need 
  not 
  

   consider 
  what 
  happen- 
  to 
  the 
  rest 
  of 
  the 
  light 
  as 
  we 
  are 
  con- 
  

   e. 
  -rued 
  only 
  with 
  what 
  is 
  called 
  the 
  ' 
  beam 
  of 
  light' 
  which 
  

   advances 
  past 
  the 
  hole. 
  

  

  68. 
  We 
  have 
  to 
  find 
  out 
  what 
  this 
  beam 
  of 
  light 
  is. 
  It 
  is 
  

   plainlv 
  no 
  longer 
  the 
  simple 
  stellade 
  which 
  encountered 
  screen 
  

   X 
  . 
  and 
  which 
  consisted 
  ol* 
  u 
  E 
  W'a 
  of 
  different 
  wave-lengths 
  

   aU 
  travelling 
  the 
  same 
  way, 
  Bui 
  it 
  is 
  light 
  of 
  some 
  kind, 
  and 
  

   as 
  Buch 
  may 
  be 
  resolved 
  into 
  u 
  £ 
  w's 
  or 
  in 
  any 
  other 
  way 
  that 
  

   i< 
  legitimate. 
  One 
  of 
  these 
  legitimate 
  resolutions 
  (and 
  the 
  

   oie- 
  which 
  will 
  most 
  assist 
  us 
  when 
  experimenting 
  with 
  an 
  

   apparatus 
  in 
  which 
  lens 
  L' 
  has 
  a 
  focal 
  length 
  less 
  than 
  the 
  

   whole 
  of 
  the 
  distance 
  from 
  that 
  lens 
  to 
  plane 
  Y') 
  is 
  a 
  resolu- 
  

   tion 
  into 
  u 
  - 
  w's 
  ( 
  undulations 
  of 
  spherical 
  wavelets) 
  which 
  

   shall 
  radiate 
  from 
  the 
  several 
  points 
  of 
  a 
  surface 
  W 
  (see 
  

   Plate 
  XII. 
  J, 
  of 
  such 
  a 
  form 
  and 
  so 
  situated 
  that 
  its 
  points 
  are 
  

   the 
  conjugate 
  foci 
  with 
  reference 
  to 
  lens 
  1/ 
  of 
  the 
  several 
  

   points 
  of 
  plane 
  Y' 
  which 
  stands 
  in 
  front 
  of 
  pro-telescope 
  T'. 
  

   The 
  positions 
  of 
  these 
  conjugate 
  foci 
  may 
  be 
  determined 
  by 
  

   Geometrical 
  Optics, 
  since 
  this 
  is 
  a 
  case 
  in 
  which 
  Geometrical 
  

   Optics 
  leads 
  to 
  a 
  correct 
  result. 
  Furthermore, 
  instead 
  of 
  

   employing 
  all 
  the 
  points 
  of 
  surface 
  W 
  as 
  the 
  centres 
  of 
  in- 
  

   numerable 
  undulations 
  of 
  infinitesimal 
  intensity, 
  we 
  may, 
  as 
  

   in 
  other 
  similar 
  cases, 
  distribute 
  a 
  large 
  but 
  finite 
  number 
  of 
  

   points 
  over 
  surface 
  W 
  at 
  sufficiently 
  short 
  intervals 
  and 
  

  

  