﻿Trowhridge 
  — 
  Investigatio7i 
  of 
  the 
  Coherer. 
  201 
  

  

  above 
  mentioned 
  : 
  for 
  each 
  capacity 
  the 
  mean 
  of 
  about 
  20 
  

   observations 
  was 
  taken 
  on 
  the 
  lower 
  limit 
  of 
  resistance 
  attained 
  

   by 
  the 
  coherer 
  on 
  discharging 
  the 
  capacity 
  through 
  it 
  ; 
  this 
  

   resistance 
  being 
  indirectly 
  obtained 
  by 
  Ohm's 
  law 
  from 
  the 
  

   readings 
  of 
  the 
  milli-ammeter. 
  

  

  As 
  I 
  have 
  mentioned 
  above, 
  the 
  upper 
  limit, 
  or 
  normal 
  

   resistance 
  of 
  the 
  coherer 
  was 
  arranged 
  to 
  have 
  a 
  uniform 
  value 
  

   of 
  about 
  2000 
  ohms. 
  

  

  ^ext, 
  the 
  charging 
  potential 
  was 
  given 
  another 
  value 
  and 
  a 
  

   like 
  set 
  of 
  observations 
  taken, 
  and 
  so 
  on 
  through 
  the 
  range 
  of 
  

   the 
  charging 
  potential. 
  

  

  It 
  was 
  at 
  once 
  evident 
  that 
  the 
  lower 
  limits 
  of 
  the 
  resistance 
  

   was 
  not 
  the 
  simple 
  function 
  of 
  the 
  quantity 
  of 
  electricity 
  sent 
  

   through 
  the 
  coherer 
  that 
  I 
  had 
  expected. 
  If 
  this 
  had 
  been 
  the 
  

   case, 
  I 
  should 
  have 
  found 
  the 
  lower 
  limit 
  of 
  the 
  resistance 
  the 
  

   same 
  for 
  the 
  cases 
  Q 
  = 
  0^^ 
  = 
  ^^2 
  ~ 
  C^gYg, 
  etc. 
  = 
  constant. 
  

   From 
  my 
  observations 
  this 
  was 
  distinctly 
  not 
  the 
  case. 
  

  

  I 
  found 
  that 
  for 
  Q 
  = 
  constant, 
  the 
  larger 
  the 
  charging 
  poten- 
  

   tial 
  — 
  and 
  hence 
  the 
  smaller 
  the 
  capacity 
  — 
  the 
  lower 
  was 
  the 
  

   final 
  resistance 
  or 
  the 
  greater 
  the 
  conductivity 
  of 
  the 
  coherer 
  

   after 
  the 
  discharge. 
  

  

  I 
  have 
  given 
  the 
  results 
  of 
  my 
  observations 
  on 
  the 
  ball 
  

   coherer 
  in 
  curves 
  1 
  and 
  2. 
  A 
  table 
  of 
  the 
  numerical 
  values 
  I 
  

   obtained 
  I 
  have 
  not 
  given, 
  since 
  it 
  would 
  be 
  of 
  little 
  theoretical 
  

   import, 
  as 
  it 
  would 
  hold 
  only 
  for 
  the 
  coherer 
  I 
  used. 
  The 
  

   form 
  of 
  the 
  curves 
  given 
  should, 
  I 
  believe, 
  be 
  similar 
  for 
  all 
  

   coherers. 
  

  

  In 
  curve 
  1, 
  the 
  abscissas 
  represent 
  the 
  reciprocals 
  of 
  the 
  lower 
  

   limit 
  of 
  resistance 
  of 
  the 
  coherer, 
  or 
  the 
  conductivity 
  ; 
  the 
  

   ordinates, 
  the 
  quantity 
  of 
  electricity 
  sent 
  through. 
  The 
  seven 
  

   different 
  curves 
  represent 
  the 
  change 
  in 
  conductivity 
  as 
  a 
  func- 
  

   tion 
  of 
  the 
  quantity 
  for 
  the 
  seven 
  different 
  charging 
  potentials 
  

   used; 
  the 
  indices 
  Y 
  = 
  10, 
  12, 
  18, 
  etc., 
  denoting 
  these 
  poten- 
  

   tials. 
  

  

  As 
  will 
  be 
  seen, 
  all 
  the 
  curves 
  tend 
  in 
  a 
  general 
  way, 
  as 
  they 
  

   should, 
  towards 
  a 
  point 
  on 
  the 
  conductivity 
  axis, 
  very 
  near 
  the 
  

   origin 
  ; 
  a 
  point 
  which 
  would 
  denote 
  the 
  reciprocal 
  of 
  2000 
  

   ohms, 
  the 
  normal 
  resistance. 
  Also 
  it 
  is 
  clear 
  that 
  the 
  greater 
  

   the 
  charging 
  potential, 
  the 
  more 
  rapid 
  the 
  rise 
  of 
  the 
  con- 
  

   ductivity 
  per 
  unit 
  increase 
  in 
  quantity 
  discharged. 
  

  

  I 
  was 
  unable, 
  with 
  the 
  coherer 
  used, 
  to 
  get 
  coherer 
  action 
  

   when 
  the 
  charging 
  potential 
  of 
  8 
  volts 
  and 
  under 
  was 
  used. 
  

   This 
  point 
  is 
  brought 
  out 
  by 
  the 
  set 
  of 
  curves 
  ^o. 
  2, 
  in 
  which 
  

   the 
  abscissas 
  are 
  the 
  same 
  as 
  in 
  curves 
  No. 
  1, 
  while 
  the 
  ordi- 
  

   nates 
  represent 
  charging 
  potentials. 
  All 
  the 
  curves 
  Q 
  = 
  con- 
  

   stant 
  cut 
  the 
  ordinate 
  axis 
  at 
  a 
  point 
  Y 
  = 
  8*75 
  volts. 
  This 
  

   is, 
  so 
  to 
  speak, 
  the 
  critical 
  voltage 
  for 
  the 
  coherer 
  under 
  

  

  