﻿Gases 
  by 
  Collision 
  with 
  Low 
  Electric 
  Forces. 
  1037 
  

  

  by 
  considering 
  the 
  distances 
  AC 
  and 
  CD 
  separately. 
  From 
  

   obseryations 
  at 
  distances 
  21 
  and 
  2, 
  b 
  electrons 
  are 
  generated 
  

   in 
  AC 
  and 
  a 
  electrons 
  in 
  CD. 
  But 
  the 
  b 
  electrons 
  generate 
  

   new 
  ones 
  in 
  passing 
  through 
  the 
  extra 
  distance 
  CD, 
  and 
  when 
  

   they 
  haye 
  reached 
  D 
  have 
  increased 
  to 
  the 
  number 
  be* 
  1 
  *. 
  

  

  Hence 
  

  

  Similarly 
  

   and 
  

  

  (1) 
  c 
  = 
  be 
  aI 
  + 
  a. 
  

  

  (2) 
  c=ae* 
  ed 
  +b 
  

  

  (3) 
  b 
  = 
  ae* 
  l 
  + 
  a. 
  

  

  The 
  yalue 
  o£ 
  a 
  may 
  be 
  found 
  from 
  either 
  of 
  these 
  three 
  

   equations, 
  two 
  of 
  which 
  are 
  independent. 
  

  

  The 
  results 
  thus 
  obtained 
  for 
  air 
  are 
  given 
  in 
  Table 
  I. 
  and 
  

   the 
  agreement 
  between 
  the 
  yalues 
  affords 
  a 
  confirmation 
  of 
  

   the 
  theory. 
  

  

  The 
  effect 
  of 
  slow 
  moving 
  electrons 
  emitted 
  from 
  the 
  

   electrodes 
  is 
  automatically 
  eliminated 
  by 
  this 
  method. 
  

  

  *l 
  P 
  =f(Xlp) 
  £or 
  Air. 
  

  

  The 
  first 
  experiment 
  was 
  to 
  determine 
  the 
  values 
  of 
  oc/p 
  

   for 
  low 
  yalues 
  of 
  X/p. 
  

  

  A 
  pressure 
  of 
  9 
  mm. 
  of 
  mercury 
  was 
  obtained 
  after 
  the 
  

   vessel 
  had 
  been 
  three 
  times 
  washed 
  out 
  with 
  air 
  thoroughly 
  

   dried, 
  and 
  exhausted 
  to 
  about 
  a 
  tenth 
  of 
  a 
  millimetre. 
  The 
  

   distances 
  between 
  the 
  plates 
  were 
  *4, 
  *8 
  and 
  1*2 
  cm. 
  and 
  the 
  

   electromotive 
  force 
  was 
  taken 
  to 
  give 
  even 
  values 
  of 
  X//?. 
  

  

  Values 
  of 
  X 
  p 
  from 
  -.10 
  to 
  100 
  were 
  thus 
  obtained. 
  

  

  Check 
  results 
  were 
  made 
  at 
  pressures 
  of 
  6 
  mm. 
  and 
  3 
  mm., 
  

   and 
  these 
  were 
  found 
  to 
  agree 
  very 
  well 
  with 
  the 
  results 
  

   obtained 
  for 
  9 
  mm. 
  

  

  Results 
  obtained 
  are 
  given 
  in 
  Table 
  I. 
  and 
  the 
  curve 
  is 
  

   shown 
  in 
  fig. 
  1 
  (PL 
  XXIV.). 
  

  

  This 
  curve 
  agrees 
  with 
  that 
  obtained 
  by 
  Townsend 
  f 
  in 
  

   ranges 
  investigated 
  by 
  him, 
  as 
  will 
  be 
  seen 
  from 
  the 
  following 
  

   table 
  : 
  — 
  

  

  X 
  

  

  60 
  

  

  70 
  

  

  . 
  

  

  90 
  

  

  1 
  

   100 
  

  

  p 
  

  

  a 
  1 
  Townsend 
  

  

  " 
  I 
  Wueatley 
  

  

  •11 
  

   118 
  

  

  •22 
  

  

  •212 
  

  

  •36 
  

   •332 
  

  

  •51 
  

   •495 
  

  

  •72 
  

  

  •733 
  ; 
  

  

  ! 
  

  

  

  It 
  was 
  found 
  that 
  for 
  values 
  of 
  X//> 
  below 
  40 
  it 
  was 
  almost 
  

   impossible 
  to 
  find 
  the 
  values 
  of 
  ajp 
  when 
  the 
  pressure 
  was 
  as 
  

  

  * 
  Townsend, 
  ' 
  Theory 
  of 
  Ionization 
  by 
  Collision, 
  

   t 
  Townsend, 
  ibid. 
  ]> 
  20. 
  

  

  a 
  2. 
  

  

  