﻿976 
  Mr. 
  R. 
  W. 
  Lawson 
  on 
  the 
  Spectra 
  of 
  

  

  to 
  test 
  the 
  discbarge 
  in 
  nitrogen 
  with 
  internal 
  electrode 
  tubes. 
  

   The 
  spectrum 
  was 
  observed 
  for 
  the 
  large 
  variety 
  of 
  cases 
  

   examined 
  with 
  the 
  external 
  electrodes, 
  and 
  found 
  to 
  be 
  

   identical 
  with 
  those. 
  The 
  bulb 
  effects, 
  however, 
  were 
  dif- 
  

   ferent, 
  and 
  behaved 
  analogously 
  with 
  those 
  of 
  the 
  ordinary 
  

   induction-coil 
  discharge. 
  In 
  one 
  instance, 
  where 
  the 
  wave- 
  

   meter 
  discharge 
  circuit 
  was 
  used, 
  and 
  the 
  diameter 
  of 
  the 
  

   capillary 
  was 
  1 
  mm., 
  the 
  current 
  passing 
  through 
  the 
  

   nitrogen 
  tube 
  with 
  aluminium 
  electrodes 
  was 
  as 
  high 
  as 
  

   0*4 
  ampere. 
  Now 
  this 
  was 
  a 
  greater 
  current 
  value 
  than 
  those 
  

   examined 
  when 
  the 
  condensed 
  discharge 
  from 
  the 
  induction 
  

   coil 
  was 
  used, 
  but 
  only 
  the 
  mixed 
  positive 
  and 
  negative 
  

   band 
  spectra 
  resulted. 
  It 
  would 
  appear 
  that 
  in 
  this 
  we 
  have 
  

   additional 
  support 
  to 
  the 
  idea 
  that 
  it 
  is 
  not 
  the 
  mean 
  value 
  

   of 
  the 
  current 
  in 
  the 
  condensed 
  discharge 
  which 
  causes 
  the 
  

   condenser 
  spectrum 
  to 
  be 
  emitted. 
  Whilst 
  the 
  mean 
  value 
  

   of 
  the 
  current 
  in 
  the 
  condenser 
  discharges 
  examined 
  was 
  

   often 
  similar 
  to 
  that 
  mentioned 
  above, 
  the 
  initial 
  value 
  at 
  

   the 
  head 
  of 
  each 
  train 
  would 
  greatly 
  exceed 
  the 
  above 
  value, 
  

   and 
  in 
  this 
  perhaps 
  we 
  have 
  the 
  reason 
  for 
  the 
  difference 
  in 
  

   spectrum 
  obtained 
  in 
  the 
  two 
  cases. 
  

  

  Air. 
  

  

  The 
  samples 
  experimented 
  on 
  were 
  freed 
  from 
  the 
  presence 
  

   of 
  carbon 
  dioxide 
  and 
  water 
  vapour 
  by 
  passing 
  over 
  caustic 
  

   potash 
  and 
  through 
  concentrated 
  sulphuric 
  acid, 
  and 
  then 
  

   allowing 
  the 
  gas 
  to 
  stand 
  over 
  phosphorus 
  pentoxide 
  till 
  

   required. 
  The 
  discharges 
  in 
  air 
  were 
  carefully 
  examined, 
  

   visually 
  and 
  photographically, 
  when 
  both 
  internal 
  and 
  ex- 
  

   ternal 
  electrodes 
  were 
  used. 
  In 
  no 
  case 
  were 
  any 
  oxygen 
  

   lines 
  discernible, 
  the 
  spectrum 
  being 
  the 
  same 
  as 
  that 
  of 
  

   pure 
  nitrogen 
  under 
  every 
  circumstance. 
  The 
  identity 
  of 
  

   the 
  spectra 
  in 
  air 
  and 
  nitrogen 
  can 
  be 
  seen 
  from 
  an 
  exami- 
  

   nation 
  of 
  spectrogram 
  4 
  (PL 
  XXII.), 
  which 
  is 
  that 
  of 
  a 
  

   Poulsen 
  discharge 
  in 
  air 
  at 
  1 
  mm. 
  pressure, 
  the 
  frequency 
  

   of 
  the 
  discharge 
  being 
  0*2 
  x 
  10 
  6 
  oscillations 
  per 
  second. 
  The 
  

   negative-glow 
  discharge 
  in 
  Geissler 
  tubes 
  with 
  internal 
  

   electrodes 
  failed 
  to 
  show 
  any 
  trace 
  of 
  the 
  spectra 
  of 
  oxygen. 
  

  

  Hydrogen, 
  

  

  This 
  was 
  prepared 
  by 
  the 
  electrolysis 
  of 
  an 
  air-free 
  solution 
  

   of 
  baryta 
  in 
  water. 
  After 
  bubbling 
  through 
  strong 
  sulphuric 
  

   acid 
  the 
  gas 
  was 
  collected 
  over 
  phosphorus 
  pentoxide, 
  as 
  in 
  

   the 
  case 
  of 
  the 
  other 
  gases. 
  

  

  At 
  constant 
  frequency 
  the 
  series 
  spectrum 
  of 
  hydrogen 
  

  

  