﻿High-Frequency 
  Discharges 
  in 
  Geissler 
  Tubes. 
  973 
  

  

  either 
  case. 
  The 
  change 
  was 
  also 
  noted 
  with 
  oxygen, 
  but 
  

   no 
  measurements 
  were 
  taken 
  in 
  this 
  case. 
  

  

  In 
  the 
  Poulsen 
  discharge 
  with 
  internal 
  electrodes, 
  the 
  

   appearance 
  of 
  the 
  bulbs 
  was 
  that 
  o£ 
  a 
  combination 
  o£ 
  the 
  

   positive 
  and 
  negative 
  bulbs 
  with 
  the 
  uncondensed 
  induction- 
  

   coil 
  discharge. 
  The 
  positive 
  column 
  extended 
  from 
  the 
  

   capillary 
  into 
  the 
  bulb, 
  and 
  at 
  the 
  higher 
  pressures 
  used 
  it 
  

   reached 
  to 
  the 
  tip 
  of 
  the 
  electrodes. 
  iSurrounding 
  each 
  

   electrode 
  was 
  the 
  negative 
  glow 
  with 
  the 
  Crookes 
  dark 
  

   space 
  of 
  normal 
  thickness. 
  The 
  negative 
  glow, 
  however, 
  

   showed 
  a 
  greater 
  proportion 
  of 
  the 
  spectrum 
  of 
  the 
  positive 
  

   column 
  than 
  was 
  the 
  case 
  with 
  the 
  ordinary 
  induction-coil 
  

   discharge. 
  This 
  combined 
  effect 
  at 
  the 
  electrodes 
  with 
  high- 
  

   frequency 
  currents 
  is 
  doubtless 
  connected 
  with 
  the 
  alternating- 
  

   nature 
  of 
  the 
  discharge, 
  the 
  electrodes 
  being 
  alternately 
  

   positive 
  and 
  negative 
  with 
  respect 
  to 
  each 
  other. 
  The 
  author 
  

   has 
  observed 
  a 
  precisely 
  similar 
  effect, 
  using 
  the 
  induction- 
  

   coil 
  discharge. 
  If 
  one 
  of 
  the 
  secondary 
  terminals 
  be 
  left 
  free, 
  

   and 
  the 
  other 
  be 
  connected 
  to 
  earth 
  through 
  the 
  Geissler 
  tube, 
  

   then 
  the 
  negative 
  glow 
  shows 
  feebly 
  round 
  both 
  electrodes 
  of 
  

   the 
  discharge 
  tube. 
  

  

  As 
  examined 
  by 
  a 
  rotating 
  mirror 
  the 
  high-frequency 
  

   Poulsen 
  discharge 
  in 
  Geissler 
  tubes 
  appeared 
  as 
  an 
  unin- 
  

   terrupted 
  band 
  of 
  light 
  of 
  uniform 
  intensity. 
  This 
  is 
  what 
  

   would 
  be 
  expected 
  with 
  such 
  a 
  type 
  of 
  undamped 
  oscillatory 
  

   discharge. 
  At 
  high 
  speeds 
  the 
  separate 
  discharges 
  were 
  

   distinctly 
  resolvable. 
  

  

  Examination 
  of 
  Spectra. 
  

  

  In 
  none 
  of 
  the 
  cases 
  examined 
  was 
  the 
  frequency 
  of 
  the 
  

   discharge 
  found 
  to 
  have 
  any 
  influence 
  on 
  the 
  spectrum 
  

   obtained, 
  this 
  being 
  the 
  same 
  at 
  lower 
  and 
  higher 
  frequencies, 
  

   other 
  conditions 
  being 
  maintained 
  constant. 
  As 
  already 
  

   mentioned, 
  the 
  limits 
  of 
  frequency 
  used 
  were 
  0*2 
  x 
  10 
  6 
  and 
  

   1*5 
  x 
  10 
  6 
  per 
  second 
  respectively. 
  Moreover, 
  in 
  none 
  of 
  

   the 
  cases 
  examined 
  was 
  it 
  found 
  possible 
  to 
  produce 
  the 
  

   condensed 
  discharge 
  spectrum 
  by 
  means 
  of 
  the 
  Poulsen 
  high- 
  

   frequency 
  generator, 
  no 
  matter 
  how 
  the 
  current 
  and 
  voltage 
  

   were 
  adjusted. 
  That 
  the 
  oscillatory 
  character 
  of 
  the 
  con- 
  

   densed 
  discharge 
  is 
  the 
  determining 
  factor 
  for 
  the 
  production 
  

   of 
  the 
  elementary 
  line-spectrum 
  of 
  a 
  gas 
  is 
  thus 
  au 
  untenable 
  

   suggestion. 
  It 
  seems 
  probable 
  that 
  the 
  production 
  of 
  the 
  

   line-spectrum 
  of 
  a 
  gas 
  by 
  use 
  of 
  condenser 
  and 
  spark-gap 
  is 
  

   dependent 
  on 
  the 
  initial 
  value 
  of 
  the 
  current 
  in 
  the 
  discharge 
  

   tube 
  at 
  each 
  train 
  of 
  oscillations. 
  If 
  this 
  be 
  the 
  case, 
  it 
  is 
  

   possible 
  that 
  with 
  very 
  heavy 
  steady 
  currents, 
  greater 
  than 
  

  

  