﻿on 
  Electric 
  Waves 
  in 
  Short 
  Wire 
  Systems. 
  597 
  

  

  It 
  might 
  certainly 
  have 
  been 
  expected 
  that 
  the 
  third 
  maximum 
  

   would 
  be 
  as 
  high 
  as 
  the 
  first 
  one 
  ; 
  it 
  is, 
  however, 
  less. 
  The 
  

   fourth 
  maximum 
  occurs 
  for 
  a 
  position 
  of 
  the 
  bridge 
  which 
  

   does 
  not 
  make 
  the 
  lengths 
  of 
  the 
  circuits 
  in 
  the 
  ratio 
  of 
  small 
  

   odd 
  numbers. 
  It 
  would 
  appear 
  here 
  that 
  a 
  comparatively 
  

   strong 
  vibration 
  of 
  twice 
  the 
  period 
  of 
  the 
  condenser 
  oscilla- 
  

   tion 
  is 
  excited 
  in 
  the 
  primary 
  circuit, 
  giving 
  a 
  stronger 
  

   impulse 
  to 
  the 
  secondary 
  circuit 
  than 
  in 
  the 
  case 
  of 
  adjacent 
  

   positions 
  of 
  the 
  bridge, 
  the 
  wave-length 
  of 
  the 
  condenser 
  

   radiation 
  being 
  9*5, 
  and 
  that 
  of 
  the 
  primary 
  circuit 
  18*88. 
  

  

  Another 
  exception 
  occurs 
  in 
  lie. 
  The 
  fifth 
  maximum 
  

   should 
  be 
  higher 
  than 
  the 
  second, 
  it 
  is 
  a 
  little 
  less. 
  For 
  the 
  fifth 
  

   maximum 
  the 
  conditions 
  are 
  favourable 
  for 
  perfect 
  resonance 
  

   between 
  the 
  primary 
  and 
  secondary 
  circuits, 
  the 
  ratio 
  of 
  their 
  

   lengths 
  being 
  as 
  1 
  to 
  1. 
  The 
  mode 
  of 
  vibration 
  of 
  the 
  

   primary 
  circuit 
  having 
  three 
  nodes 
  gives 
  a 
  wave-length 
  14*03 
  

   rather 
  nearer 
  to 
  that 
  of 
  the 
  condenser 
  radiation 
  15*2 
  than 
  

   13*86, 
  the 
  wave-length 
  of 
  the 
  mode 
  of 
  vibration 
  having 
  five 
  

   nodes 
  in 
  the 
  case 
  of 
  the 
  second 
  maximum. 
  With 
  these 
  two 
  

   exceptions, 
  the 
  explanation 
  seems 
  to 
  account 
  for 
  all 
  the 
  results 
  

   which 
  have 
  been 
  obtained. 
  The 
  question 
  of 
  the 
  relative 
  

   heights 
  of 
  the 
  maxima 
  was 
  not 
  considered 
  when 
  the 
  observa- 
  

   tions 
  were 
  taken, 
  so 
  no 
  special 
  efforts 
  were 
  made 
  to 
  keep 
  the 
  

   character 
  of 
  the 
  spark 
  absolutely 
  unaltered. 
  

  

  The 
  explanation 
  depends 
  on 
  the 
  assumption 
  that 
  the 
  

   vibrations 
  in 
  the 
  primary 
  circuit 
  are 
  forced 
  and 
  not 
  free 
  

   vibrations 
  of 
  the 
  system. 
  A 
  similar 
  explanation 
  probably 
  holds 
  

   in 
  the 
  case 
  of 
  the 
  beautiful 
  curves 
  given 
  by 
  Rubens 
  (Wied. 
  

   Ann. 
  xlii. 
  p. 
  154, 
  1891). 
  If 
  the 
  vibrations 
  of 
  the 
  primary 
  

   circuit 
  in 
  Ruben 
  s's 
  experiments 
  are 
  free 
  and 
  not 
  forced 
  by 
  

   the 
  vibrator, 
  it 
  is 
  difficult 
  to 
  see 
  why 
  any 
  other 
  mode 
  of 
  

   vibration 
  should 
  occur 
  than 
  the 
  fundamental 
  one. 
  If 
  the 
  

   primary 
  circuit 
  always 
  vibrates 
  in 
  its 
  fundamental 
  mode, 
  

   those 
  positions 
  of 
  the 
  bridge 
  which 
  divide 
  the 
  wires 
  into 
  

   circuits 
  requiring 
  this 
  mode 
  for 
  perfect 
  resonance 
  between 
  

   them 
  should 
  give 
  higher 
  maxima 
  than 
  other 
  positions. 
  This 
  

   is 
  not 
  the 
  case. 
  On 
  the 
  other 
  hand, 
  J. 
  J. 
  Thomson's 
  argument 
  

   in 
  discussing 
  these 
  experiments 
  (' 
  Recent 
  Researches/ 
  § 
  388), 
  

   showing 
  that 
  the 
  vibrations 
  in 
  the 
  wires 
  which 
  are 
  detected 
  

   by 
  the 
  bolometer 
  cannot 
  be 
  forced 
  by 
  the 
  vibrator, 
  is 
  not 
  

   sound. 
  Thomson 
  assumes 
  that 
  if 
  the 
  vibrations 
  are 
  forced, 
  

   no 
  capacity 
  being 
  at 
  the 
  ends 
  of 
  the 
  wires 
  and 
  no 
  bridge 
  

   placed 
  across 
  them, 
  the 
  places 
  where 
  the 
  electromotive 
  

   intensity 
  at 
  right 
  angles 
  to 
  the 
  wire 
  vanishes 
  are 
  an 
  odd 
  

   number 
  of 
  quarter-wave-lengths 
  of 
  the 
  condenser 
  radiation 
  

   from 
  the 
  ends 
  of 
  the 
  wires. 
  This 
  is 
  incorrect, 
  as 
  shown 
  by 
  

   the 
  results 
  of 
  the 
  experiments 
  described 
  in 
  the 
  early 
  part 
  of 
  

   this 
  paper. 
  

  

  