﻿the 
  Photoelectric 
  Effect. 
  565 
  

  

  An 
  examination 
  of 
  the 
  curves 
  for 
  sodium 
  shows 
  that 
  the 
  

   first 
  maximum 
  (a) 
  decreases 
  more 
  rapidly 
  as 
  " 
  fatigue 
  " 
  

   proceeds 
  than 
  does 
  the 
  second 
  maximum 
  (b). 
  This 
  is 
  what 
  

   we 
  should 
  expect, 
  since 
  the 
  electrons 
  liberated 
  by 
  light 
  in 
  

   the 
  region 
  of 
  the 
  second 
  maximum 
  possess 
  much 
  greater 
  

   initial 
  velocities, 
  and 
  are 
  therefore 
  better 
  able 
  to 
  penetrate 
  

   the 
  surface 
  layer 
  of 
  oxide. 
  Evidently, 
  if 
  measurements 
  

   could 
  have 
  been 
  made 
  immediately 
  after 
  the 
  sodium 
  had 
  

   been 
  distilled, 
  the 
  sensitiveness 
  at 
  (a) 
  would 
  have 
  been 
  con- 
  

   siderably 
  greater 
  than 
  at 
  (b). 
  

  

  In 
  the 
  case 
  of 
  platinum, 
  fig. 
  4, 
  both 
  maxima 
  lie 
  beyond 
  

   the 
  region 
  in 
  which 
  measurements 
  are 
  possible. 
  The 
  value 
  

   of 
  X 
  , 
  about 
  280 
  jjl/ul, 
  agrees 
  well 
  with 
  the 
  value 
  270 
  /x^ 
  

   calculated 
  theoretically. 
  

  

  In 
  fig. 
  5 
  for 
  aluminium 
  the 
  first 
  maximum 
  only 
  appears. 
  

   It 
  is 
  at 
  a 
  wave-length 
  X 
  max 
  . 
  = 
  220//,/£, 
  whereas 
  the 
  theory 
  

   would 
  put 
  it 
  at 
  240 
  //,/x. 
  A 
  possible 
  explanation 
  of 
  this 
  dis- 
  

   crepancy 
  is 
  suggested 
  by 
  the 
  fact 
  that 
  the 
  maximum 
  shifts 
  

   to 
  the 
  right 
  as 
  photoelectric 
  fatigue 
  proceeds, 
  as 
  is 
  seen 
  by 
  

   comparing 
  curves 
  1 
  and 
  3. 
  This 
  is 
  consistent 
  with 
  the 
  fact, 
  

   mentioned 
  previously, 
  that 
  metal 
  surfaces 
  become 
  more 
  

   electronegative 
  and 
  assume 
  properties 
  of 
  more 
  electronegative 
  

   metals 
  as 
  fatigue 
  takes 
  place. 
  Since 
  the 
  aluminium 
  was 
  in 
  

   contact 
  with 
  the 
  air 
  for 
  a 
  short 
  while 
  before 
  making 
  the 
  

   test, 
  it 
  is 
  certain 
  that 
  the 
  true 
  maximum 
  for 
  aluminium 
  is 
  at 
  

   a 
  wave-length 
  longer 
  than 
  220 
  /jl/jl. 
  Whether 
  the 
  entire 
  

   discrepancy 
  may 
  be 
  explained 
  in 
  this 
  way 
  is 
  not 
  known. 
  

   The 
  value 
  of 
  X 
  , 
  360 
  /jl/ul, 
  agrees 
  exactly 
  with 
  that 
  calculated 
  

   from 
  the 
  theory. 
  

  

  The 
  fatigue 
  in 
  the 
  case 
  of 
  sodium, 
  fig. 
  5, 
  was 
  less 
  than 
  in 
  

   the 
  case 
  of 
  aluminium, 
  because 
  much 
  greater 
  pains 
  were 
  

   taken 
  to 
  secure 
  a 
  good 
  vacuum. 
  There 
  is 
  no 
  certain 
  evidence 
  

   of 
  any 
  shift 
  of 
  the 
  maxima 
  with 
  time. 
  The 
  first 
  maximum 
  

   (a) 
  is 
  at 
  wave-length 
  360 
  yu,//,, 
  which 
  agrees 
  almost 
  exactly 
  

   with 
  the 
  theory. 
  The 
  second 
  maximum 
  (b) 
  is 
  at 
  wave-length 
  

   227 
  fjbfjb. 
  Because 
  of 
  the 
  wide 
  slit 
  and 
  strip, 
  and 
  consequent 
  

   overlapping 
  in 
  the 
  visible 
  spectrum, 
  X 
  cannot 
  be 
  accurately 
  

   determined 
  from 
  these 
  curves, 
  and 
  it 
  appears 
  larger 
  than 
  it 
  

   really 
  is. 
  It 
  is 
  probably 
  not 
  far 
  from 
  the 
  calculated 
  value 
  

   X 
  = 
  550/u,/x. 
  The 
  large 
  overlapping 
  in 
  this 
  spectral 
  region 
  

   is 
  shown 
  in 
  fig. 
  8 
  (p. 
  566), 
  which 
  represents 
  the 
  dispersive 
  

   power 
  of 
  the 
  instrument 
  at 
  different 
  wave-lengths. 
  There 
  

   is 
  very 
  little 
  overlapping 
  of 
  spectral 
  lines 
  for 
  wave-lengths 
  

   less 
  than 
  300 
  fjifi. 
  But 
  the 
  overlapping 
  is 
  very 
  considerable 
  

   in 
  the 
  visible 
  spectrum. 
  

  

  Phil. 
  Maq. 
  S. 
  6. 
  Yol. 
  26. 
  No. 
  154. 
  Oct. 
  1913. 
  2 
  Q 
  

  

  