﻿40 
  Prof. 
  Chattock 
  and 
  Mr. 
  Tyndall 
  on 
  Changes 
  

  

  doing 
  more 
  for 
  what 
  were 
  to 
  be 
  trial 
  readings 
  only, 
  and 
  this 
  

   turns 
  out 
  to 
  have 
  been 
  fortunate. 
  

  

  For 
  examination 
  of 
  the 
  nitrogen 
  curves 
  suggests 
  somewhat 
  

   unexpectedly 
  that 
  the 
  effects 
  observed 
  were 
  due 
  not 
  so 
  much 
  

   to 
  the 
  nitrogen 
  as 
  to 
  the 
  trace 
  of 
  oxygen 
  still 
  present 
  ; 
  at 
  any 
  

   rate 
  in 
  the 
  case 
  of 
  the 
  lower 
  percentages. 
  

  

  It 
  will 
  be 
  seen 
  that 
  the 
  positive 
  curve 
  starts 
  at 
  practically 
  

   the 
  same 
  point 
  as 
  that 
  for 
  oxygen, 
  and 
  that 
  the 
  negative 
  rises 
  

   to 
  a 
  peak 
  which 
  is 
  about 
  as 
  high 
  as 
  the 
  oxygen 
  peak. 
  It 
  is 
  

   true 
  the 
  nitrogen 
  peak 
  is 
  much 
  more 
  spread 
  out 
  along 
  the 
  

   scale 
  of 
  percentages, 
  but 
  that 
  is 
  of 
  course 
  because 
  the 
  nitrogen 
  

   percentages 
  are 
  plotted. 
  We 
  have 
  only 
  to 
  suppose 
  that 
  the 
  

   nitrogen 
  contains 
  between 
  0*1 
  and 
  02 
  per 
  cent, 
  of 
  oxygen 
  to 
  

   make 
  the 
  two 
  peaks 
  correspond. 
  

  

  Another 
  fact 
  which 
  supports 
  the 
  oxygen 
  theory 
  is 
  the 
  ease 
  

   with 
  which 
  we 
  were 
  able 
  to 
  clear 
  out 
  the 
  active 
  impurity. 
  

   At 
  a 
  rate 
  of 
  discharge 
  which 
  would 
  have 
  required 
  many 
  

   weeks 
  to 
  clear 
  out 
  2 
  per 
  cent, 
  of 
  oxygen, 
  we 
  were 
  able 
  in 
  

   a 
  few 
  hours 
  to 
  reduce 
  the 
  same 
  percentage 
  of 
  nitrogen 
  to 
  

   inactivity. 
  

  

  On 
  this 
  assumption 
  — 
  that 
  the 
  contractions 
  observed 
  when 
  

   nitrogen 
  is 
  the 
  impurity 
  are 
  really 
  due 
  to 
  traces 
  of 
  oxygen 
  — 
  

   we 
  have 
  calculated 
  the 
  amount 
  of 
  this 
  oxygen 
  which 
  must 
  

   have 
  been 
  present 
  ; 
  expressing 
  it 
  as 
  a 
  percentage 
  of 
  the 
  

   nitrogen, 
  and 
  using 
  of 
  course 
  the 
  ratio 
  2' 
  8 
  of 
  Table 
  2. 
  

  

  The 
  only 
  percentages 
  for 
  which 
  the 
  discharge 
  happens 
  to 
  

   have 
  been 
  continued 
  to 
  the 
  drop 
  are 
  the 
  first 
  five 
  ; 
  and 
  these 
  

   are 
  given 
  in 
  Table 
  3, 
  along 
  with 
  the 
  amount 
  of 
  oxygen 
  

   calculated. 
  

  

  Table 
  3. 
  

  

  Percentage 
  of 
  

   Nitrogen. 
  

  

  Percentage 
  of 
  Oxygen 
  

   in 
  the 
  Nitrogen. 
  

  

  0-0014 
  

   •010 
  

   •075 
  

   •206 
  

  

  222 
  

  

  8-5 
  

   11 
  

  

  •77 
  

   •40 
  

   •17 
  

  

  These 
  numbers 
  not 
  only 
  indicate 
  the 
  exhausting 
  of 
  a 
  

   limited 
  supply 
  of 
  the 
  active 
  agent, 
  whatever 
  it 
  is, 
  but 
  they 
  

  

  