﻿578 
  On 
  Ionization 
  and 
  Wireless 
  Telegraphy 
  . 
  

  

  the 
  air 
  in 
  the 
  zinc-plate 
  condenser 
  was 
  ionized. 
  A 
  displace- 
  

   ment 
  of 
  one 
  centimetre 
  o£ 
  the 
  cylindrical 
  condenser 
  of 
  the 
  

   cymometer 
  corresponds 
  to 
  a 
  change 
  of 
  4*54 
  E.S. 
  C.G.S. 
  units 
  

   in 
  the 
  capacity 
  of 
  the 
  zinc-plate 
  condenser 
  which, 
  in 
  its 
  

   normal 
  state, 
  had 
  a 
  capacity 
  of 
  2099 
  E.S. 
  C.G.S. 
  units. 
  This 
  

   shift 
  of 
  one 
  centimetre 
  accordingly 
  compensates 
  a 
  fractional 
  

   change 
  of 
  O0022 
  in 
  apparent 
  capacity 
  or 
  in 
  the 
  square 
  of 
  the 
  

   frequency 
  of 
  the 
  circuit. 
  

  

  The 
  experimental 
  results 
  are 
  also 
  plotted 
  as 
  graphs 
  in 
  

   tigs. 
  8 
  and 
  9, 
  in 
  which 
  the 
  thicknesses 
  of 
  plates 
  penetrated 
  by 
  

   the 
  ionizing 
  agents 
  are 
  the 
  abscissae 
  and 
  the 
  compensating 
  

   shifts 
  on 
  the 
  cymometer 
  are 
  the 
  ordinates. 
  

  

  In 
  comparing 
  the 
  theoretically 
  derived 
  curve 
  shown 
  in 
  

   fig. 
  2 
  with 
  the 
  experimental 
  graphs, 
  these 
  latter 
  should 
  be 
  

   rotated 
  in 
  their 
  own 
  plane 
  through 
  180°. 
  Their 
  ordinates 
  

   have 
  then 
  the 
  correct 
  algebraic 
  signs 
  for 
  this 
  comparison, 
  

   and 
  the 
  greater 
  ionizations 
  (obtained 
  by 
  penetration 
  through 
  

   thinner 
  plates) 
  are 
  to 
  the 
  right 
  also. 
  But 
  absolute 
  agree- 
  

   ment 
  cannot 
  be 
  expected, 
  for 
  in 
  the 
  theoretical 
  curve 
  the 
  

   leakances 
  are 
  proportional 
  to 
  the 
  abscissae, 
  and 
  this 
  cannot 
  

   be 
  asserted 
  of 
  the 
  experimental 
  graphs, 
  even 
  when 
  in 
  the 
  

   rotated 
  position. 
  

  

  Roman 
  numerals 
  are 
  inserted 
  in 
  the 
  right-hand 
  column 
  

   of 
  the 
  table. 
  Each 
  such 
  numeral 
  serves 
  to 
  identify 
  the 
  series 
  

   of 
  observations 
  given 
  on 
  that 
  line. 
  Each 
  graph 
  is 
  based 
  on 
  

   the 
  observations 
  occupying 
  a 
  single 
  line, 
  and, 
  in 
  the 
  figure, 
  

   the 
  graph 
  bears 
  the 
  numeral 
  by 
  which 
  the 
  observations 
  are 
  

   characterized 
  in 
  the 
  table. 
  

  

  It 
  may 
  be 
  noted 
  that 
  the 
  results 
  given 
  in 
  the 
  table 
  are 
  

   based 
  on 
  nearly 
  nine 
  hundred 
  electrometer 
  throws. 
  

  

  The 
  curves 
  show 
  that 
  the 
  separate 
  observations 
  in 
  each 
  

   series 
  were 
  fairly 
  consistent 
  but 
  that 
  the 
  ionizations 
  differed 
  

   from 
  one 
  series 
  to 
  another, 
  so 
  that 
  the 
  separate 
  curves 
  occur 
  

   in 
  different 
  positions. 
  

  

  Further, 
  the 
  curves 
  follow 
  the 
  general 
  form 
  theoretically 
  

   derived, 
  but 
  several 
  show 
  positive 
  ordinates 
  for 
  the 
  thickest 
  

   plates 
  instead 
  of 
  very 
  small 
  negative 
  ones 
  as 
  theory 
  would 
  

   indicate. 
  It 
  appears 
  that, 
  from 
  some 
  unknown 
  cause, 
  these 
  

   experimental 
  curves 
  are 
  shifted 
  up 
  somewhat 
  from 
  the 
  

   position 
  theoretically 
  deduced. 
  But 
  considering 
  the 
  smallness 
  

   of 
  the 
  effect 
  to 
  be 
  measured, 
  the 
  agreement 
  is 
  perhaps 
  as 
  

   close 
  as 
  could 
  be 
  expected. 
  

  

  University 
  College, 
  Nottingham, 
  

   July 
  5, 
  1913. 
  

  

  