﻿the 
  Photoelectric 
  Effect. 
  553 
  

  

  taken 
  from 
  our 
  last 
  paper. 
  Considering 
  the 
  fact 
  that 
  the 
  

   intrinsic 
  potentials 
  o£ 
  the 
  investigated 
  surfaces 
  may 
  have 
  

   been 
  somewhat 
  different 
  from 
  the 
  values 
  given 
  in 
  the 
  tables, 
  

   and 
  since 
  the 
  finite 
  size 
  of 
  the 
  spectrometer-slits 
  caused 
  a 
  

   slight 
  overlapping 
  in 
  the 
  spectrum, 
  these 
  results 
  are 
  in 
  very 
  

   fair 
  support 
  of 
  the 
  theory. 
  

  

  We 
  have, 
  therefore, 
  good 
  experimental 
  support 
  of 
  those 
  

   results 
  of 
  the 
  theory 
  expressed 
  by 
  

  

  N 
  = 
  0, 
  when 
  <: 
  hv 
  < 
  w 
  

  

  e 
  

  

  and 
  Tr 
  300 
  7 
  , 
  300 
  7 
  

  

  V 
  = 
  Iiv—Wq, 
  when 
  wn-< 
  hv 
  < 
  0. 
  

  

  e 
  e 
  

  

  There 
  still 
  remains 
  to 
  be 
  tested 
  the 
  equation 
  

  

  AT 
  Ah 
  /., 
  e 
  w 
  \ 
  , 
  300 
  7 
  

  

  N= 
  BVV 
  1 
  "S00^> 
  when^^ 
  — 
  A,<0. 
  . 
  (b) 
  

  

  The 
  purpose 
  of 
  this 
  paper 
  is 
  the 
  experimental 
  investigation 
  

   of 
  the 
  relations 
  involved 
  in 
  this 
  equation. 
  

  

  Preliminary 
  Considerations. 
  

  

  At 
  the 
  outset 
  it 
  should 
  be 
  remarked 
  that 
  equation 
  (6) 
  is 
  

   not 
  necessarily 
  the 
  only 
  or 
  the 
  complete 
  expression 
  of 
  the 
  

   theory. 
  The 
  thermodynamic 
  and 
  statistical 
  treatment 
  yielded 
  

   two 
  integral 
  equations 
  which 
  were 
  not 
  satisfied 
  by 
  any 
  

   analytic 
  function 
  continuous 
  throughout 
  the 
  entire 
  range 
  

   of 
  variation. 
  It 
  was 
  found, 
  however, 
  that 
  equations 
  (3) 
  and 
  

   (I) 
  are 
  a 
  solution 
  of 
  these 
  integral 
  equations. 
  The 
  complete 
  

   solution 
  may 
  therefore 
  consist 
  of 
  the 
  sum 
  of 
  several 
  terms, 
  

   of 
  which 
  equation 
  (6) 
  is 
  only 
  one 
  part. 
  This 
  suggestion 
  will 
  

   be 
  considered 
  again 
  later 
  in 
  the 
  paper. 
  It 
  is 
  also 
  possible 
  

   that 
  the 
  approximations 
  used 
  lead 
  to 
  au 
  error 
  in 
  this 
  part 
  of 
  

   the 
  solution. 
  

  

  One 
  evident 
  requirement 
  of 
  the 
  theory 
  is 
  that 
  the 
  rate 
  of 
  

   photoelectric 
  emission 
  must 
  reach 
  a 
  maximum 
  at 
  some 
  par- 
  

   ticular 
  frequency, 
  instead 
  of 
  increasing 
  indefinitely 
  with 
  the 
  

   frequency 
  as 
  was 
  at 
  first 
  supposed. 
  For 
  from 
  equation 
  (6) 
  

  

  dN 
  __ 
  Ah 
  / 
  e 
  3wq 
  J\ 
  _ 
  n 
  

  

  whence 
  

  

  dv 
  KVV300 
  hv 
  

   3 
  e 
  tc 
  Q 
  _ 
  3 
  

  

  rmas 
  -~2 
  300 
  h 
  ~ 
  2 
  V 
  °' 
  ' 
  [i) 
  

  

  and 
  v 
  min 
  .=:cc 
  (8) 
  

  

  <:/ 
  2 
  X 
  . 
  ( 
  negative 
  when 
  v 
  = 
  ^v 
  

   dv^ 
  \. 
  positive 
  when 
  v= 
  x> 
  

  

  