﻿Line 
  Spectrum 
  from 
  Uncharged 
  Molecules. 
  907 
  

  

  connexions 
  to 
  the 
  dynamo 
  could 
  be 
  reversed, 
  so 
  that 
  F 
  

   could 
  be 
  charged 
  either 
  positively 
  or 
  negatively. 
  

  

  In 
  the 
  tabe 
  in 
  which 
  measurements 
  were 
  made, 
  the 
  

   diameter 
  of 
  the 
  opening 
  at 
  c 
  was 
  0*8 
  cm. 
  and 
  at 
  F 
  was 
  

   2 
  8 
  cm. 
  The 
  distance 
  from 
  F 
  to 
  the 
  middle 
  of 
  T 
  was 
  

   14 
  cm. 
  The 
  gauze 
  had 
  5*6 
  wires 
  per 
  cm. 
  For 
  starting 
  

   the 
  arc. 
  a 
  tube 
  (not 
  shown 
  in 
  the 
  diagram) 
  extended 
  from 
  

   into 
  a 
  mercury 
  cup 
  which 
  could 
  be 
  raised 
  and 
  lowered 
  so 
  as 
  

   to 
  cause 
  the 
  mercury 
  in 
  C 
  to 
  make 
  and 
  break 
  connexion 
  

   with 
  that 
  in 
  B. 
  

  

  If 
  the 
  arc 
  is 
  started 
  in 
  such 
  a 
  tube, 
  the 
  luminosity 
  gra- 
  

   dually 
  passes 
  up 
  towards 
  F. 
  If 
  F 
  is 
  charged 
  positively 
  

   to 
  20 
  or 
  SO 
  volts 
  above 
  the 
  potential 
  of 
  the 
  middle 
  of 
  the 
  

   arc 
  at 
  the 
  time 
  that 
  the 
  luminous 
  vapour 
  begins 
  to 
  pass 
  c, 
  

   a 
  discharge 
  passes 
  to 
  F 
  and 
  striations 
  appear 
  below 
  F. 
  

   There 
  is 
  then 
  a 
  non-luminous 
  region 
  above 
  c 
  corresponding 
  

   to 
  the 
  dark 
  space 
  near 
  the 
  cathode 
  in 
  an 
  ordinary 
  vacuum 
  

   discharge. 
  As 
  the 
  luminous 
  vapour 
  rises, 
  the 
  striations 
  rise 
  

   and 
  disappear 
  as 
  they 
  come 
  to 
  F. 
  Whm 
  the 
  last 
  one 
  dis- 
  

   appears, 
  the 
  region 
  between 
  F 
  and 
  c 
  is 
  dark 
  when 
  F 
  is 
  

   charged 
  and 
  luminous 
  when 
  F 
  is 
  not 
  charged. 
  

  

  If 
  the 
  potential 
  difference 
  between 
  F 
  and 
  the 
  middle 
  of 
  

   the 
  arc 
  is 
  at 
  this 
  time 
  more 
  than 
  15 
  or 
  20 
  volts, 
  there 
  is 
  an 
  

   increase 
  in 
  brightness 
  at 
  c. 
  If 
  the 
  potential 
  difference 
  is 
  

   somewhat 
  smaller 
  than 
  this, 
  the 
  luminosity 
  between 
  F 
  and 
  c 
  

   will 
  still 
  be 
  destroyed 
  but 
  there 
  will 
  be 
  no 
  increase 
  in 
  

   brightness 
  at 
  c. 
  It 
  is 
  this 
  latter 
  condition 
  which 
  gives 
  us 
  

   the 
  most 
  knowledge 
  concerning 
  the 
  nature 
  of 
  the 
  light 
  

   from 
  the 
  vapour, 
  and 
  will 
  be 
  discussed 
  in 
  the 
  following- 
  

   paragraphs. 
  

  

  It 
  should 
  be 
  noted 
  that 
  while 
  it 
  is 
  not 
  difficult 
  to 
  secure 
  

   the 
  right 
  conditions 
  for 
  this 
  experiment, 
  it 
  is 
  difficult 
  to 
  

   maintain 
  them 
  for 
  any 
  length 
  of 
  time. 
  As 
  the 
  tube 
  becomes 
  

   heated, 
  greater 
  quantities 
  of 
  vapour 
  pass 
  into 
  the 
  condensing 
  

   chamber 
  and 
  the 
  region 
  becomes 
  more 
  luminous. 
  If 
  at 
  such 
  

   a 
  time 
  F 
  is 
  charged 
  to 
  8 
  or 
  10 
  volts, 
  the 
  light 
  is 
  not 
  all 
  

   destroyed. 
  If 
  it 
  is 
  charged 
  to 
  more 
  than 
  this, 
  the 
  luminosity 
  

   is 
  increased 
  instead 
  of 
  being 
  diminished. 
  This 
  increase 
  in 
  

   light 
  is 
  no 
  doubt 
  the 
  result 
  of 
  ionization 
  between 
  F 
  and 
  c, 
  

   which 
  is 
  caused 
  by 
  the 
  increase 
  in 
  the 
  electric 
  field, 
  as 
  

   is 
  shown 
  by 
  the 
  increase 
  in 
  the 
  conductivity 
  of 
  the 
  vapour. 
  

   In 
  previous 
  work 
  performed 
  by 
  myself 
  this 
  region 
  was 
  

   examined 
  while 
  in 
  such 
  a 
  condition, 
  and 
  as 
  a 
  result 
  I 
  came 
  

   to 
  the 
  erroneous 
  conclusion 
  that 
  the 
  luminosity 
  could 
  not 
  be 
  

   destroyed 
  by 
  an 
  electric 
  field 
  *. 
  

  

  * 
  Phvs. 
  Eev. 
  xxii. 
  p. 
  221 
  (1906). 
  

  

  