﻿924 
  A 
  Special 
  Case 
  of 
  Gaseous 
  Conduction. 
  

  

  gas. 
  But 
  the 
  existence 
  of 
  the 
  effect 
  does 
  not 
  seem 
  to 
  

   invalidate 
  the 
  main 
  conclusion 
  of 
  this 
  paper, 
  that 
  the 
  form 
  

   of 
  conduction 
  observed 
  is 
  one 
  in 
  which 
  the 
  diffusion 
  of 
  the 
  

   ions 
  is 
  overwhelmingly 
  important 
  as 
  compared 
  with 
  their 
  

   recombination. 
  This 
  conclusion 
  does 
  not 
  affect 
  in 
  any 
  way 
  

   our 
  theory 
  of 
  conduction 
  ; 
  the 
  prediction 
  of 
  such 
  a 
  form 
  is 
  

   implicit 
  in 
  the 
  accepted 
  theory, 
  but 
  it 
  does 
  not 
  seem 
  to 
  have 
  

   been 
  uoted 
  that 
  the 
  necessary 
  conditions 
  are 
  capable 
  of 
  

   experimental 
  realization. 
  

  

  Summary* 
  

  

  § 
  1, 
  2. 
  A 
  theory 
  of 
  conduction 
  in 
  an 
  ionized 
  gas 
  is 
  sketched 
  

   applicable 
  to 
  experimental 
  conditions, 
  such 
  that 
  the 
  effects 
  

   of 
  the 
  recombination 
  of 
  the 
  ions 
  are 
  inappreciable 
  compared 
  

   with 
  those 
  of 
  their 
  diffusion. 
  It 
  appears 
  that 
  if 
  such 
  conditions 
  

   could 
  be 
  realized 
  the 
  form 
  of 
  the 
  tc 
  saturation 
  curve 
  " 
  of 
  the 
  

   ionized 
  gas 
  should 
  depend 
  on 
  nothing 
  but 
  the 
  temperature 
  of 
  

   the 
  gas. 
  

  

  § 
  3-8. 
  The 
  results 
  are 
  given 
  of 
  some 
  recent 
  measurements 
  

   of 
  the 
  " 
  saturation 
  curves" 
  of 
  air, 
  hydrogen, 
  and 
  carbon 
  

   dioxide 
  at 
  pressures 
  less 
  than 
  atmospheric 
  contained 
  between 
  

   plates 
  less 
  than 
  3 
  mm. 
  apart, 
  and 
  ionized 
  by 
  a 
  or 
  X 
  rays. 
  

   It 
  appears 
  incidentally 
  that 
  metal 
  electrodes 
  do 
  not 
  always 
  

   absorb 
  all 
  the 
  ions 
  falling 
  on 
  them. 
  

  

  § 
  9, 
  10. 
  It 
  is 
  shown 
  that 
  the 
  form 
  of 
  these 
  saturation 
  

   curves 
  is 
  inexplicable 
  on 
  the 
  theory 
  of 
  conduction 
  which 
  

   applies 
  to 
  cases 
  where 
  recombination 
  is 
  the 
  important 
  process. 
  

   On 
  the 
  other 
  hand, 
  this 
  form 
  is 
  almost 
  exactly 
  that 
  given 
  by 
  

   the 
  theory 
  sketched 
  in 
  §2; 
  it 
  is 
  found, 
  as 
  it 
  is 
  predicted, 
  

   that 
  the 
  form 
  of 
  the 
  saturation 
  curve 
  is, 
  within 
  certain 
  limits, 
  

   independent 
  of 
  the 
  nature 
  and 
  pressure 
  of 
  the 
  gas, 
  of 
  the 
  

   distance 
  between 
  the 
  electrodes, 
  and 
  of 
  the 
  intensity 
  of 
  the 
  

   ionization. 
  These 
  limits 
  are 
  probably 
  set, 
  on 
  the 
  one 
  hand, 
  

   by 
  the 
  influence 
  of 
  recombination, 
  on 
  the 
  other, 
  by 
  the 
  finite 
  

   initial 
  velocities 
  of 
  the 
  S 
  rays 
  emitted 
  in 
  ionization. 
  

  

  § 
  11. 
  Certain 
  anomalies 
  were 
  found 
  when 
  hard 
  /3 
  and 
  y 
  rays 
  

   were 
  used 
  for 
  ionization 
  instead 
  of 
  a 
  and 
  X 
  rays. 
  These 
  

   anomalies 
  are 
  not 
  explained 
  ; 
  they 
  are 
  connected 
  with 
  actions 
  

   taking 
  place 
  at 
  the 
  surface 
  of 
  the 
  electrodes, 
  but 
  their 
  presence 
  

   does 
  not 
  seem 
  to 
  affect 
  the 
  conclusions 
  attained 
  in 
  the 
  rest 
  of 
  

   the 
  paper. 
  

  

  