﻿Anomalous 
  Conduction 
  in 
  a 
  Solid 
  Dielectric. 
  693 
  

  

  On 
  this 
  view, 
  as 
  the 
  field 
  is 
  increased, 
  the 
  average 
  value 
  of 
  

   cos 
  0, 
  and 
  consequently 
  of 
  the 
  current, 
  might 
  be 
  expected 
  

   to 
  increase 
  at 
  first 
  in 
  proportion 
  to 
  the 
  field. 
  As 
  the 
  field 
  

   was 
  further 
  increased, 
  cos 
  might 
  be 
  expected 
  to 
  vary 
  more 
  

   rapidly 
  with 
  the 
  field 
  than 
  is 
  required 
  by 
  Ohm's 
  law, 
  and 
  in 
  

   a 
  manner 
  such 
  as 
  our 
  experiments 
  suggest, 
  finally 
  attaining 
  

   saturation 
  for 
  extremely 
  large 
  fields. 
  The 
  quantity 
  p 
  repre- 
  

   senting 
  the 
  rate 
  of 
  expulsion 
  of 
  electrons 
  from 
  the 
  atoms 
  

   would 
  also 
  vary 
  with 
  the 
  field, 
  because 
  as 
  the 
  doublet 
  axes 
  

   become 
  more 
  and 
  more 
  parallel 
  to 
  the 
  field, 
  they 
  become 
  

   more 
  favourably 
  situated 
  in 
  respect 
  to 
  each 
  other 
  for 
  the 
  

   purpose 
  of 
  pulling 
  out 
  corpuscles. 
  

  

  It 
  may 
  be 
  remarked, 
  that 
  if 
  the 
  above 
  theory 
  were 
  true, 
  

   we 
  should 
  expect 
  a 
  variation 
  of 
  the 
  specific 
  inductive 
  capacity 
  

   with 
  the 
  field. 
  The 
  fact 
  that 
  the 
  value 
  of 
  the 
  specific 
  

   inductive 
  capacity 
  between 
  plates 
  A 
  and 
  C 
  would 
  be 
  greater 
  

   than 
  that 
  between 
  B 
  and 
  C 
  would 
  tend 
  to 
  diminish 
  the 
  rise 
  

   in 
  potential 
  of 
  the 
  plate 
  A, 
  and 
  the 
  fall 
  of 
  potential 
  of 
  B, 
  

   but 
  it 
  is 
  easy 
  to 
  see 
  that 
  this 
  effect 
  does 
  not 
  completely 
  

   counteract 
  nor 
  reverse 
  the 
  alteration 
  in 
  potentials 
  of 
  the 
  

   plates. 
  

  

  Summary. 
  

  

  (1) 
  Experiments 
  for 
  obtaining 
  information 
  with 
  regard 
  

   to 
  the 
  current 
  potential 
  curve 
  for 
  a 
  very 
  good 
  insulator 
  are 
  

   described, 
  the 
  method 
  being 
  of 
  a 
  new 
  type 
  designed 
  to 
  avoid 
  

   difficulties 
  due 
  to 
  the 
  high 
  potentials 
  used, 
  and 
  to 
  bring 
  into 
  

   prominence 
  any 
  deviation 
  from 
  Ohm's 
  law. 
  

  

  (2) 
  The 
  theory 
  and 
  possibilities 
  of 
  error 
  due 
  to 
  a 
  departure 
  

   from 
  the 
  ideal 
  conditions 
  of 
  the 
  simple 
  theory 
  are 
  fully 
  

   considered, 
  and 
  it 
  is 
  shown 
  that 
  the 
  only 
  likely 
  way 
  of 
  

   accounting 
  for 
  the 
  results 
  obtained 
  is 
  by 
  the 
  assumption 
  of 
  a 
  

   departure 
  from 
  Ohm's 
  law 
  such 
  as 
  corresponds 
  to 
  an 
  increase 
  

   of 
  the 
  conductivity 
  with 
  the 
  field. 
  

  

  (3) 
  The 
  reasons 
  for 
  discarding 
  any 
  explanation 
  of 
  the 
  

   phenomenon 
  as 
  due 
  to 
  soakage 
  are 
  discussed, 
  and 
  the 
  

   possibility 
  of 
  its 
  explanation 
  as 
  the 
  result 
  of 
  an 
  increase 
  in 
  

   the 
  ionization, 
  due 
  to 
  the 
  field, 
  is 
  examined 
  and 
  found 
  

   untenable. 
  

  

  Other 
  methods 
  of 
  accounting 
  for 
  the 
  phenomenon 
  are 
  also 
  

   discussed, 
  the 
  most 
  likely 
  one 
  being 
  framed 
  on 
  the 
  lines 
  of 
  a 
  

   modification 
  of 
  the 
  theory 
  of 
  metallic 
  conduction 
  proposed 
  

   by 
  Sir 
  J. 
  J. 
  Thomson. 
  

  

  The 
  Physical 
  Laboratory, 
  

  

  University 
  of 
  Sheffield, 
  

   April 
  21st, 
  1913. 
  

  

  Phil. 
  Mag. 
  S, 
  6. 
  Vol. 
  26. 
  No. 
  154. 
  Oct. 
  1913, 
  3 
  A 
  

  

  