﻿972 
  Dr. 
  G. 
  Johnstone 
  Stoncy 
  on 
  

  

  figure) 
  which 
  the 
  centres 
  of 
  their 
  concentration 
  images 
  reach. 
  

   The 
  partials 
  belonging 
  to 
  class 
  /■, 
  if 
  all 
  the 
  rest 
  were 
  excluded, 
  

   would 
  acting 
  together 
  form 
  an 
  exceedingly 
  good 
  image 
  at 
  

   a?', 
  since 
  the 
  partials 
  of 
  this 
  group 
  are 
  all 
  optimum 
  partials. 
  

   This 
  is 
  a 
  -rate 
  of 
  things 
  which 
  the 
  micro 
  scopist 
  can 
  realise 
  

   by 
  closing 
  the 
  iris-diaphragm 
  onder 
  his 
  condenser 
  sufficiently 
  

   to 
  exclude 
  the 
  light 
  which 
  would 
  develop 
  any 
  other 
  partials 
  

   than 
  those 
  of 
  class 
  k. 
  Tin's 
  in 
  using 
  the 
  microscope 
  is 
  called 
  

   'clearing 
  the 
  image.' 
  We 
  can 
  do 
  the 
  same 
  in 
  the 
  author's 
  

   experimental 
  apparatus, 
  by 
  calling 
  into 
  requisition 
  an 
  iris- 
  

   diaphragm 
  provided 
  at 
  F 
  (see 
  Plate 
  XII.). 
  Hut 
  the 
  

   astronomer 
  oas 
  no 
  means 
  of 
  accomplishing 
  it 
  with 
  his 
  tele- 
  

   scope. 
  He 
  must 
  put 
  up 
  with 
  all 
  those 
  mischievous 
  had 
  

   partials, 
  being 
  present 
  in 
  the 
  image 
  which 
  the 
  telescope 
  

   presents 
  to 
  him. 
  Ami 
  under 
  the 
  conditions 
  represented 
  in 
  

   6g, 
  2, 
  l>\ 
  they 
  contribute 
  so 
  largely 
  to 
  the 
  image 
  that 
  the 
  

   resulting 
  ' 
  complete 
  imam-' 
  cannot 
  yield 
  a 
  really 
  good 
  image 
  

   of 
  the 
  outline 
  of 
  Neptune, 
  which 
  is 
  whai 
  we 
  want 
  to 
  see, 
  

   although 
  the 
  astr< 
  nomer. 
  would 
  probably 
  consider 
  it 
  quite 
  a 
  

   fair 
  image. 
  This 
  is 
  partly 
  because 
  of 
  a 
  very 
  curious 
  power 
  

   which 
  the 
  experienced 
  observer 
  acquires 
  of, 
  in 
  such 
  cases, 
  

   being 
  able 
  in 
  some 
  degree 
  to 
  detect 
  the 
  image 
  which 
  the 
  

   better 
  of 
  the 
  contributory 
  partials 
  would 
  Eorm, 
  and 
  being 
  able 
  

   to 
  distinguish 
  and 
  exclude 
  from 
  his 
  consideration 
  some 
  at 
  

   least 
  o\' 
  the 
  added 
  defects. 
  

  

  By 
  the 
  movements 
  provided 
  at 
  the 
  B 
  end 
  of 
  the 
  board 
  BE, 
  

   the 
  observer 
  can 
  bring 
  the 
  concentration 
  image 
  successively 
  

   into 
  all 
  the 
  positions 
  described 
  in 
  the 
  last 
  paragraph, 
  and 
  can, 
  

   then 
  see 
  for 
  himself 
  the 
  partial 
  images 
  in 
  the 
  pro-telescope, 
  

   which 
  will 
  result 
  ; 
  and 
  the 
  astronomer 
  would 
  do 
  well 
  to. 
  

   familiarise 
  himself 
  with 
  these 
  details. 
  

  

  85. 
  We 
  have 
  now 
  to 
  contrast 
  with 
  the 
  above 
  the 
  image 
  

   which 
  the 
  same 
  telescope 
  will 
  furnish 
  of 
  an 
  object 
  of 
  the 
  same 
  

   shape 
  but 
  of 
  larger 
  angular 
  size, 
  when 
  the 
  feature 
  to 
  be 
  

   seen 
  is 
  the 
  outline 
  of 
  the 
  disk 
  of 
  planet 
  Mars. 
  Here 
  when 
  

   we 
  distribute 
  the 
  partials 
  into 
  the 
  four 
  groups 
  k 
  I 
  m 
  and 
  n, 
  

   we 
  find 
  this 
  distribution 
  represented 
  by 
  fig. 
  2, 
  b" 
  . 
  This 
  

   diagram 
  shows 
  us 
  that 
  the 
  optimum 
  partials, 
  which 
  are 
  those 
  

   of 
  group 
  k, 
  are 
  greatly 
  more 
  numerous* 
  than 
  they 
  were 
  

   when 
  Neptune 
  was 
  the 
  object, 
  and 
  at 
  the 
  same 
  time 
  the 
  

   partials 
  of 
  the 
  three 
  other 
  groups, 
  I 
  m 
  and 
  n, 
  are 
  much 
  

   fewer. 
  Hence 
  the 
  ' 
  complete 
  image 
  ' 
  of 
  the 
  outline 
  of 
  the 
  

   planet 
  Mars 
  which 
  will 
  result 
  from 
  their 
  being 
  all 
  present, 
  

  

  * 
  The 
  numbers 
  of 
  the 
  partials 
  "of 
  each 
  group 
  are 
  proportional 
  to 
  the- 
  

   areas 
  allotted 
  to 
  them 
  in 
  figs. 
  2, 
  3 
  and 
  4. 
  The 
  intensities 
  of 
  the 
  partials. 
  

   have 
  however 
  to 
  be 
  taken 
  into 
  account 
  as 
  well 
  as 
  their 
  number. 
  

  

  