﻿554: 
  Mr. 
  K. 
  T. 
  Compton 
  and 
  Prof. 
  0. 
  W. 
  Richardson 
  on 
  

  

  Therefore 
  equation 
  (6) 
  requires 
  the 
  photoelectric 
  sensitiveness 
  

   of 
  any 
  metal 
  to 
  be 
  a 
  maximum 
  at 
  a 
  wave-length 
  equal 
  to 
  irX 
  , 
  

   where 
  \ 
  is 
  the 
  longest 
  wave-length 
  at 
  which 
  electrons 
  are 
  

   emitted, 
  and 
  it 
  requires 
  that 
  the 
  sensitiveness 
  approach 
  zero 
  

   when 
  the 
  wave-length 
  is 
  exceedingly 
  small. 
  

  

  The 
  existence 
  o£ 
  a 
  maximum 
  in 
  the 
  curve 
  representing 
  the 
  

   relation 
  between 
  frequency 
  and 
  rate 
  of 
  emission 
  is 
  proved 
  

   by 
  the 
  work 
  done 
  on 
  a 
  large 
  number 
  of 
  metals 
  by 
  Pohl 
  and 
  

   Pringsheim 
  *. 
  The 
  results 
  of 
  their 
  experiments 
  may 
  be 
  

   generalized 
  in 
  the 
  statement 
  that 
  the 
  wave-length-rate 
  of 
  

   emission 
  curve 
  for 
  every 
  metal 
  is 
  of 
  the 
  type 
  shown 
  in 
  

   fig. 
  1, 
  abed 
  e, 
  and 
  that 
  for 
  each 
  metal 
  the 
  starting-point 
  and 
  

   the 
  position 
  of 
  the 
  maximum 
  are 
  at 
  wave-lengths 
  peculiar 
  to 
  

   the 
  given 
  metal. 
  N 
  is 
  the 
  number 
  of 
  electrons 
  emitted 
  

  

  Fig.l. 
  

  

  per 
  second 
  by 
  a 
  unit 
  amount 
  of 
  light-energy. 
  They 
  finu 
  

   that 
  the 
  part 
  b 
  c 
  d 
  of 
  the 
  curve 
  is 
  present 
  only 
  when 
  there 
  is 
  

   a 
  component 
  of 
  the 
  electric 
  intensity 
  in 
  the 
  light- 
  wave 
  which 
  

   is 
  perpendicular 
  to 
  the 
  mirror 
  surface 
  of 
  the 
  metal. 
  When 
  

   the 
  light 
  is 
  incident 
  perpendicularly 
  the 
  part 
  bed 
  degenerates 
  

   into 
  the 
  dotted 
  line 
  b 
  d, 
  which 
  forms 
  part 
  of 
  a 
  continuous 
  

   curve 
  abdg. 
  This 
  curve 
  is 
  not 
  in 
  good 
  agreement 
  with 
  

   equation 
  (6), 
  which 
  is 
  represented 
  by 
  the 
  curve 
  a/eg. 
  

   However, 
  the 
  theoretical 
  curve 
  afcg 
  was 
  obtained 
  on 
  the 
  

   supposition 
  that 
  the 
  radiation 
  incident 
  on 
  the 
  metal 
  is 
  isotropic, 
  

   while 
  the 
  experimental 
  curve 
  of 
  Pohl 
  and 
  Pringsheim 
  was 
  

   obtained 
  with 
  the 
  light 
  incident 
  at 
  a 
  certain 
  definite 
  angle. 
  

   They 
  found 
  that 
  the 
  relative 
  importance 
  of 
  the 
  " 
  hump 
  " 
  b 
  c 
  d 
  

   depends 
  greatly 
  on 
  the 
  angle 
  of 
  incidence. 
  Therefore 
  these 
  

   curves 
  are 
  not 
  strictly 
  comparable. 
  

  

  However, 
  Pohl 
  and 
  Pringsheim 
  found 
  that 
  the 
  position 
  of 
  

   the 
  maximum 
  is 
  nearly 
  or 
  entirely 
  independent 
  of 
  the 
  angle 
  

  

  * 
  Verh. 
  d. 
  D. 
  Phijs. 
  Ges. 
  vol. 
  xiii. 
  p. 
  474 
  (1911) 
  ; 
  vol. 
  xii. 
  p. 
  349 
  

   (1910) 
  ; 
  vol. 
  xii. 
  p. 
  682 
  (1910) 
  ; 
  vol. 
  xiv. 
  p. 
  46 
  (1912), 
  &c. 
  

  

  