﻿[ 
  678 
  ] 
  

  

  LV. 
  A 
  Case 
  of 
  Anomalous 
  Conduction 
  in 
  a 
  Solid 
  Dielectric. 
  

   By 
  W. 
  F. 
  Gr. 
  Swank, 
  D.Sc, 
  A.R.C.S., 
  Assistant 
  Lecturer 
  

   in 
  Physics 
  at 
  the 
  University 
  of 
  Sheffield 
  *'. 
  

  

  (1) 
  TI^HE 
  conductivity 
  produced 
  by 
  the 
  action 
  of 
  X 
  rays 
  in 
  

   JL 
  dielectrics 
  has 
  been 
  investigated 
  by 
  several 
  observers. 
  

   The 
  present 
  paper, 
  however, 
  deals 
  with 
  the 
  natural 
  con- 
  

   ductivity 
  of 
  paraffin-wax 
  when 
  under 
  the 
  influence 
  of 
  no 
  

   external 
  agency 
  except 
  the 
  electric 
  field 
  responsible 
  for 
  the 
  

   current. 
  The 
  experiments 
  suggest 
  that 
  Ohm's 
  law 
  is 
  not 
  

   obeyed, 
  the 
  conductivity 
  increasing 
  in 
  fact 
  with 
  the 
  field. 
  

   The 
  fact 
  is 
  of 
  interest 
  because 
  it 
  has 
  an 
  important 
  bearing 
  

   on 
  the 
  mechanism 
  of: 
  the 
  conduction 
  ; 
  but 
  these 
  theoretical 
  

   considerations 
  will 
  be 
  dealt 
  with 
  after 
  the 
  experiments 
  have 
  

   been 
  described. 
  

  

  The 
  natural 
  conductivity 
  of 
  paraffin- 
  wax 
  being 
  so 
  small, 
  

   we 
  are 
  able, 
  and 
  indeed 
  we 
  are 
  obliged, 
  to 
  employ 
  potential 
  

   gradients 
  enormously 
  greater 
  than 
  those 
  employed 
  in 
  the 
  

   case 
  of 
  good 
  conductors. 
  The 
  potential 
  gradients 
  employed 
  

   in 
  the 
  present 
  work 
  are 
  of 
  the 
  order 
  50,000 
  volts 
  per 
  

   centimetre. 
  

  

  At 
  first 
  sight, 
  the 
  simplest 
  way 
  to 
  investigate 
  the 
  relation 
  

   between 
  the 
  current 
  and 
  the 
  potential 
  gradient 
  would 
  be 
  

   to 
  place 
  the 
  wax 
  between 
  two 
  plates, 
  one 
  connected 
  to 
  an 
  

   electrometer 
  and 
  the 
  other 
  kept 
  at 
  a 
  potential 
  of 
  say 
  

   10,000 
  volts, 
  and 
  to 
  note 
  the 
  rate 
  of 
  leak 
  into 
  the 
  electro- 
  

   meter. 
  Apart 
  from 
  the 
  difficulty 
  of 
  obtaining 
  a 
  battery 
  

   giving 
  10,000 
  volts, 
  however, 
  the 
  slightest 
  fluctuation 
  in 
  the 
  

   potential 
  of 
  the 
  battery 
  and 
  consequently 
  of 
  the 
  plate 
  con- 
  

   nected 
  to 
  it 
  would, 
  owing 
  to 
  inductive 
  action 
  on 
  the 
  plate 
  

   connected 
  to 
  the 
  electrometer, 
  cause 
  movements 
  of 
  the 
  

   needle 
  such 
  as 
  to 
  entirely 
  mask 
  the 
  movements 
  due 
  to 
  

   the 
  conduction 
  of 
  the 
  electricity 
  across 
  the 
  wax. 
  Again, 
  in 
  

   order 
  to 
  avoid 
  trouble 
  due 
  to 
  leakage 
  through 
  the 
  air, 
  it 
  

   is 
  necessary 
  that 
  the 
  charged 
  plate 
  should 
  be 
  embedded 
  

   entirely 
  in 
  the 
  wax 
  ; 
  and 
  even 
  then 
  it 
  is 
  necessary 
  that 
  the 
  

   wax 
  faces 
  remote 
  from 
  the 
  charged 
  plate 
  should 
  be 
  kept 
  at 
  

   zero 
  potential, 
  as, 
  if 
  any 
  part 
  is 
  exposed, 
  it 
  is 
  liable 
  to 
  get 
  

   up 
  to 
  a 
  high 
  potential 
  and 
  cause 
  slow 
  discharge 
  from 
  

   the 
  air. 
  

  

  The 
  apparatus 
  represented 
  in 
  fig. 
  1 
  was 
  devised 
  to 
  overcome 
  

   these 
  effects, 
  and 
  it 
  further 
  had 
  the 
  advantage, 
  as 
  will 
  be 
  seen 
  

   later, 
  of 
  not 
  indicating 
  the 
  part 
  of 
  the 
  current 
  obeying 
  Ohm's 
  

   law, 
  but 
  only 
  the 
  part 
  corresponding 
  to 
  the 
  deviation 
  from 
  

  

  * 
  Communicated 
  by 
  the 
  Author. 
  

  

  