﻿980 
  Spectra 
  of 
  High-Frequency 
  Discharges 
  in 
  Geissler 
  Tubes. 
  

  

  in 
  spectrum. 
  It 
  was 
  found, 
  however, 
  that 
  by 
  instantaneously 
  

   reducing 
  the 
  current 
  the 
  red 
  spectrum 
  was 
  produced 
  even 
  in 
  

   the 
  hot 
  tube, 
  so 
  that 
  bulk 
  temperature 
  has 
  little, 
  i£ 
  any, 
  in- 
  

   fluence. 
  This 
  direct 
  transition 
  from 
  the 
  red 
  to 
  the 
  blue 
  

   spectrum 
  is 
  shown 
  remarkably 
  well 
  in 
  spectrograms 
  6 
  a, 
  6 
  b, 
  

   & 
  6 
  c 
  (PL 
  XXII. 
  ), 
  which 
  were 
  all 
  taken 
  at 
  the 
  same 
  frequency, 
  

   and 
  correspond 
  to 
  current 
  strengths 
  indicated. 
  It 
  will 
  be 
  

   noticed 
  that 
  the 
  red 
  spectrum 
  shows 
  feebly 
  even 
  on 
  6 
  c. 
  This 
  

   is 
  due 
  to 
  the 
  occasional 
  fall 
  of 
  the 
  current 
  in 
  the 
  discharge 
  

   circuit, 
  owing 
  to 
  slight 
  unsteadiness 
  in 
  the 
  arc 
  generator. 
  As 
  

   observed 
  visually, 
  the 
  transformation 
  to 
  blue 
  spectrum 
  ap- 
  

   peared 
  complete. 
  The 
  change 
  from 
  6 
  a 
  to 
  6 
  c 
  is 
  seen 
  to 
  be 
  

   gradual, 
  and 
  the 
  correspondence 
  between 
  6 
  c 
  and 
  the 
  blue 
  

   spectrum 
  is 
  clear 
  from 
  comparison 
  with 
  6 
  d, 
  which 
  is 
  a 
  

   photograph 
  of 
  the 
  blue 
  spectrum 
  as 
  obtained 
  by 
  using 
  a 
  

   condensed 
  induction-coil 
  discharge 
  at 
  practically 
  the 
  same 
  

   mean 
  current 
  density. 
  As 
  mentioned 
  above, 
  it 
  thus 
  appears 
  

   likely 
  that 
  it 
  is 
  the 
  initial 
  value 
  of 
  the 
  discharge 
  in 
  each 
  

   train 
  rather 
  than 
  the 
  mean 
  value 
  of 
  the 
  current 
  which 
  deter- 
  

   mines 
  the 
  spectrum 
  produced. 
  Hence, 
  since 
  the 
  complete 
  

   transformation 
  from 
  the 
  red 
  to 
  the 
  blue 
  spectrum 
  was 
  obtained 
  

   by 
  use 
  of 
  a 
  constant-frequency 
  undamped 
  discharge 
  with 
  

   increasing 
  current 
  values, 
  we 
  can 
  but 
  conclude 
  that 
  current 
  

   density 
  is 
  the 
  main 
  determining 
  influence. 
  In 
  the 
  case 
  of 
  

   the 
  condenser 
  discharge, 
  the 
  maximum 
  value 
  of 
  the 
  current 
  

   would 
  greatly 
  exceed 
  the 
  mean 
  value 
  indicated, 
  but 
  the 
  effect 
  

   would 
  be 
  to 
  produce 
  only 
  more 
  definitely 
  the 
  blue 
  spectrum. 
  

   It 
  should 
  be 
  mentioned 
  that 
  the 
  low-current 
  red 
  spectrum, 
  

   as 
  obtained 
  with 
  the 
  Poulsen 
  discharge, 
  is 
  identical 
  in 
  the 
  

   minutest 
  detail 
  with 
  that 
  obtained 
  with 
  the 
  ordinary 
  induction- 
  

   coil 
  discharge. 
  The 
  fainter 
  upper 
  portions 
  of 
  the 
  spectro- 
  

   grams 
  are 
  of 
  the 
  bulb 
  of 
  the 
  discharge 
  tube, 
  and 
  in 
  all 
  cases 
  

   showed 
  the 
  red 
  spectrum 
  — 
  a 
  result 
  adding 
  further 
  confirma- 
  

   tion 
  to 
  the 
  above 
  conclusion 
  as 
  to 
  the 
  effect 
  of 
  current 
  density. 
  

   Close 
  up 
  to 
  the 
  electrodes 
  both 
  the 
  white 
  and 
  blue 
  spectra 
  

   were 
  obtained, 
  the 
  latter 
  never 
  perfectly 
  pure. 
  

  

  Neon, 
  Helium, 
  and 
  Carbon 
  Dioxide. 
  

  

  Standard 
  tubes 
  of 
  these 
  gases 
  were 
  examined 
  under 
  the 
  

   variety 
  of 
  cases 
  used 
  with 
  other 
  gases, 
  but 
  in 
  every 
  case 
  the 
  

   spectrum 
  obtained 
  was 
  identical 
  with 
  that 
  obtained 
  when 
  

   using 
  the 
  induction-coil 
  discharge 
  without 
  condenser. 
  Fre- 
  

   quency-change 
  was 
  wholly 
  without 
  influence. 
  The 
  diameter 
  

   of 
  the 
  capillary 
  of 
  the 
  neon 
  tube 
  was 
  2 
  mm., 
  and 
  the 
  maximum 
  

   current 
  used 
  was 
  0*09 
  ampere 
  ; 
  for 
  the 
  helium 
  tube 
  these 
  

  

  