﻿36 
  Prof. 
  Chattock 
  and 
  Mr. 
  Tyndall 
  on 
  Changes 
  

  

  using 
  point 
  discharge 
  in 
  C0 
  2 
  at 
  atmospheric 
  pressure, 
  under 
  

   conditions 
  which 
  were 
  therefore 
  much 
  more 
  nearly 
  comparable 
  

   with 
  ours, 
  found 
  that 
  at 
  least 
  3 
  per 
  cent, 
  of 
  the 
  C0 
  2 
  could 
  

   be 
  dissociated, 
  and 
  that 
  at 
  this 
  point 
  marked 
  re-association 
  

   was 
  occurring. 
  

  

  In 
  order 
  therefore 
  to 
  prove 
  that 
  the 
  drop 
  in 
  our 
  curves 
  

   corresponds 
  with 
  an 
  equilibrium 
  point 
  for 
  which 
  the 
  pressure 
  

   of 
  the 
  oxygen 
  is 
  small, 
  it 
  is 
  necessary 
  to 
  show 
  that 
  discharge 
  

   through 
  hydrogen 
  containing 
  water-vapour 
  only 
  does 
  not 
  

   result 
  in 
  much 
  dissociation 
  of 
  the 
  latter. 
  This 
  is 
  the 
  point 
  

   we 
  have 
  not 
  yet 
  been 
  able 
  to 
  test 
  directly, 
  but 
  some 
  light 
  is 
  

   obtainable 
  on 
  it 
  by 
  an 
  examination 
  of 
  Curves 
  V. 
  

  

  The 
  water-vapour 
  formed 
  in 
  the 
  discharge 
  vessel 
  accu- 
  

   mulates 
  there, 
  as 
  it 
  has 
  to 
  diffuse 
  through 
  about 
  18 
  inches 
  of 
  

   narrow 
  quill-tubing 
  before 
  reaching 
  the 
  P 
  2 
  Os 
  — 
  a 
  process 
  

   which 
  must 
  require 
  many 
  hours. 
  Hence 
  the 
  drop 
  following 
  

   upon 
  the 
  clearing 
  out 
  of 
  a 
  large 
  percentage 
  of 
  oxygen 
  occurs 
  

   in 
  the 
  presence 
  of 
  more 
  water-vapour 
  than 
  that 
  following 
  a 
  

   small 
  percentage. 
  It 
  will 
  be 
  noticed, 
  however, 
  that 
  in 
  every 
  

   case 
  where 
  the 
  level 
  part 
  of 
  the 
  curve 
  is 
  able 
  to 
  form, 
  the 
  

   experimental 
  points 
  rise 
  to 
  the 
  same 
  high 
  value 
  just 
  before 
  

   dropping. 
  The 
  horizontal 
  line 
  AB 
  makes 
  this 
  clear 
  *. 
  

  

  Now 
  if 
  we 
  may 
  take 
  this 
  as 
  an 
  indication 
  that 
  the 
  samo 
  

   amount 
  of 
  oxygen 
  is 
  present 
  in 
  every 
  case 
  just 
  before 
  the 
  drop, 
  

   it 
  will 
  follow 
  either 
  that 
  the 
  drop 
  is 
  not 
  an 
  equilibrium 
  point 
  

   at 
  all, 
  or 
  else 
  that 
  when 
  it 
  is 
  reached 
  the 
  amount 
  of 
  oxygen 
  

   has 
  become 
  very 
  small 
  ; 
  otherwise 
  it 
  will 
  be 
  impossible 
  for 
  

   the 
  small 
  amount 
  of 
  subsequent 
  combination 
  indicated 
  by 
  the 
  

   later 
  contractions 
  to 
  bring 
  about 
  equilibrium 
  between 
  the 
  

   equal 
  amounts 
  of 
  oxygen 
  and 
  the 
  very 
  unequal 
  amounts 
  of 
  

   water- 
  vapour 
  in 
  the 
  hydrogen. 
  

  

  Curves 
  V. 
  may 
  be 
  regarded 
  from 
  a 
  second 
  point 
  of 
  view. 
  

   They 
  are 
  so 
  nearly 
  level 
  up 
  to 
  the 
  drop 
  that 
  equal 
  distances 
  

   along 
  the 
  time 
  axis 
  mean 
  nearly 
  equal 
  amounts 
  of 
  oxygen 
  

   removed. 
  In 
  other 
  words, 
  the 
  abscissas 
  up 
  to, 
  but 
  not 
  much 
  

   beyond 
  the 
  drop, 
  give 
  percentages 
  of 
  oxygen 
  (proportionally) 
  to 
  

   within 
  a 
  constant 
  when 
  measured 
  from 
  right 
  to 
  left. 
  The 
  drop 
  

   curves 
  are 
  in 
  fact 
  approximately 
  the 
  peak 
  plotted 
  backwards. 
  

   In 
  Curves 
  VIII. 
  (second 
  air) 
  is 
  given 
  a 
  magnified 
  drawing 
  

   of 
  the 
  peak 
  obtained 
  for 
  the 
  same 
  conditions 
  as 
  Curves 
  V., 
  

   except 
  that 
  the 
  gas 
  was 
  dry 
  and 
  the 
  current 
  was 
  larger. 
  On 
  

   the 
  assumption 
  that 
  practically 
  the 
  whole 
  of 
  the 
  oxygen 
  is 
  gone 
  

   when 
  the 
  drop 
  is 
  reached, 
  we 
  have 
  constructed 
  a 
  curve 
  from 
  

  

  * 
  Curve 
  /, 
  it 
  is 
  true, 
  does 
  not 
  rise 
  quite 
  so 
  high 
  as 
  as 
  the 
  others, 
  but 
  

   this 
  is 
  doubtless 
  because 
  the 
  discharges 
  in 
  this 
  case 
  were 
  kept 
  on 
  for 
  so 
  

   many 
  minutes 
  that 
  the 
  last 
  reading 
  is 
  the 
  average 
  of 
  the 
  highest 
  with 
  

   lower 
  parts 
  of 
  the 
  curve, 
  and 
  is 
  therefore 
  too 
  low. 
  

  

  