﻿Distribution 
  of 
  Energy 
  in 
  tlie 
  Spectra 
  of 
  Gases. 
  827 
  

  

  order 
  to 
  account 
  for 
  the 
  proportionality 
  between 
  radiation 
  

   and 
  current-density 
  we 
  have 
  only 
  to 
  suppose 
  that 
  the 
  

   number 
  of 
  ions 
  formed 
  per 
  c.c. 
  per 
  sec. 
  is 
  proportional 
  to 
  

   the 
  current, 
  a 
  supposition 
  which 
  meets 
  with 
  no 
  difficulty 
  in 
  

   the 
  theory 
  of 
  the 
  passage 
  of 
  electricity 
  through 
  gases. 
  The 
  

   fact 
  that 
  absorption 
  can 
  be 
  observed 
  in 
  a 
  straight 
  tube 
  with 
  

   no 
  apparent 
  differences 
  in 
  intensity 
  can 
  be 
  explained 
  by 
  sup- 
  

   posing 
  that 
  there 
  are 
  places 
  where 
  most 
  of 
  the 
  ions 
  have 
  

   come 
  from 
  elsewhere, 
  moving 
  wdth 
  high 
  velocities, 
  and 
  thus 
  

   maintaining 
  the 
  necessary 
  current-density 
  with 
  only 
  a 
  few 
  

   ions 
  per 
  c.c. 
  Or 
  if 
  the 
  continuous 
  spectrum 
  mentioned 
  in 
  

   this 
  paper 
  above 
  came 
  from 
  different 
  centres 
  of 
  emission 
  

   from 
  the 
  bright 
  lines 
  this 
  would 
  account 
  for 
  the 
  absorption 
  

   observed. 
  

  

  Summary 
  and 
  Conclusion. 
  

  

  The 
  results 
  of 
  the 
  present 
  investigations 
  may 
  be 
  summarized 
  

   as 
  follows 
  :■ 
  — 
  

  

  (1) 
  If 
  the 
  current 
  in 
  a 
  discharge-tube 
  be 
  small, 
  the 
  radia- 
  

   tion 
  from 
  it 
  is 
  proportional 
  to 
  the 
  current-density, 
  both 
  for 
  

   the 
  whole 
  spectrum 
  and 
  for 
  any 
  portion 
  of 
  it. 
  

  

  (2) 
  In 
  heavy 
  discharges 
  such 
  as 
  that 
  produced 
  by 
  a 
  large 
  

   condenser 
  in 
  parallel 
  with 
  the 
  discharge-tube, 
  there 
  is 
  strong 
  

   evidence 
  that 
  the 
  energy 
  shifts 
  towards 
  the 
  longer 
  wave- 
  

   lengths 
  when 
  the 
  current-density 
  increases. 
  Whether 
  the 
  

   energy 
  of 
  any 
  portion 
  of 
  the 
  spectrum 
  increases 
  at 
  a 
  rate 
  

   greater 
  or 
  smaller 
  than 
  the 
  current 
  has 
  not 
  been 
  detei 
  mined, 
  

   owing 
  to 
  the 
  difficulties 
  in 
  the 
  way 
  of 
  measuring 
  these 
  

   currents. 
  

  

  (3) 
  When 
  the 
  pressure 
  at 
  which 
  a 
  condensed 
  discbarge 
  is 
  

   sent 
  through 
  the 
  tube 
  is 
  increased, 
  the 
  energy 
  shifts 
  towards 
  

   the 
  longer 
  wave-lengths. 
  A 
  continuous 
  spectrum 
  makes 
  its 
  

   appearance 
  in 
  these 
  discharges, 
  and 
  becomes 
  more 
  marked, 
  

   the 
  higher 
  the 
  pressure. 
  

  

  The 
  above 
  investigations 
  make 
  no 
  claim 
  to 
  completeness, 
  

   and 
  confirmation 
  of 
  them 
  is 
  eminently 
  desirable. 
  The 
  writer 
  

   hopes 
  to 
  continue 
  the 
  research 
  and 
  to 
  extend 
  it 
  to 
  other 
  gases. 
  

   To 
  Prof. 
  Sir 
  Joseph 
  Thomson, 
  for 
  suggesting 
  the 
  research, 
  

   and 
  for 
  much 
  interest 
  and 
  advice, 
  he 
  owes 
  his 
  best 
  thanks. 
  

   Prof. 
  Paschen 
  of 
  Tubingen 
  has 
  helped 
  the 
  work 
  on 
  much 
  

   by 
  his 
  kindness 
  in 
  demonstrating 
  the 
  principles 
  of 
  his 
  

   galvanometer 
  and 
  in 
  providing 
  one 
  of 
  these 
  out 
  of 
  his 
  own 
  

   laboratory. 
  

  

  