﻿L. 
  Page— 
  The 
  Photoelectric 
  Effect. 
  505 
  

  

  once. 
  In 
  no 
  case 
  was 
  any 
  difference 
  found 
  in 
  the 
  maximum 
  

   velocity 
  as 
  thus 
  determined 
  and 
  the 
  maximum 
  velocity 
  as 
  

   found 
  by 
  the 
  method 
  described 
  above. 
  

  

  In 
  figures 
  2-7 
  are 
  plotted 
  the 
  photoelectric 
  currents 
  against 
  

   the 
  accelerating 
  (positive) 
  potentials 
  in 
  volts 
  applied 
  to 
  the 
  case 
  

   and 
  receiving 
  electrode. 
  The 
  scale 
  of 
  ordinates 
  is 
  arbitrary, 
  

   though 
  the 
  ordinates 
  of 
  the 
  " 
  scraped 
  " 
  and 
  " 
  unscraped 
  " 
  

   curves 
  for 
  \ 
  = 
  2190 
  are 
  in 
  the 
  proper 
  ratio. 
  Figures 
  2, 
  4, 
  and 
  

   6 
  refer 
  to 
  the 
  aluminium 
  receiving 
  electrode, 
  and 
  figures 
  3, 
  5, 
  

   and 
  7 
  to 
  the 
  copper 
  receiving 
  electrode. 
  The 
  full 
  line 
  refers 
  

   to 
  wave 
  length 
  2190, 
  and 
  the 
  broken 
  line 
  to 
  wave 
  length 
  2160. 
  

  

  Headings 
  were 
  taken 
  for 
  each 
  metal 
  with 
  each 
  receiving 
  

   electrode, 
  both 
  after 
  scraping 
  the 
  metal, 
  and 
  after 
  having 
  ex- 
  

   posed 
  the 
  metal 
  to 
  the 
  air 
  for 
  fifteen 
  hours 
  or 
  more. 
  In 
  the 
  

   figures 
  the 
  latter 
  are 
  marked 
  " 
  unscraped." 
  

  

  Discussion 
  of 
  Curves. 
  

  

  Comparison 
  of 
  figures 
  2 
  and 
  3, 
  4 
  and 
  5, 
  6 
  and 
  7 
  shows 
  that 
  

   the 
  only 
  effect 
  of 
  replacing 
  the 
  aluminium 
  receiving 
  electrode 
  

   "by 
  the 
  copper 
  one 
  is 
  to 
  shift 
  the 
  curves 
  to 
  the 
  right 
  relative 
  to 
  

   the 
  current 
  axis 
  by 
  0'9 
  volt. 
  This 
  is 
  quite 
  exactly 
  the 
  case 
  for 
  

   the 
  unscraped 
  metals, 
  and 
  also 
  for 
  the 
  Aluminium-Aluminium 
  

   and 
  Aluminium-Copper 
  curves 
  of 
  the 
  scraped 
  metals. 
  In 
  the 
  

   other 
  cases 
  of 
  scraped 
  metals 
  the 
  point 
  of 
  saturation 
  is 
  not 
  so 
  

   well 
  defined, 
  though 
  the 
  point 
  where 
  the 
  curve 
  meets 
  the 
  volt- 
  

   age 
  axis 
  shows 
  a 
  shift 
  of 
  very 
  closely 
  0*9 
  volt. 
  On 
  account 
  of 
  

   the 
  impossibility 
  of 
  always 
  scraping 
  to 
  the 
  same 
  degree, 
  and 
  

   on 
  account 
  of 
  the 
  rapid 
  fatigue 
  immediately 
  after 
  scraping, 
  

   which 
  was 
  particularly 
  pronounced 
  in 
  the 
  case, 
  of 
  zinc, 
  it 
  was 
  

   difficult 
  to 
  locate 
  very 
  precisely 
  the 
  saturation 
  points 
  in 
  the 
  

   case 
  of 
  the 
  scraped 
  metals. 
  Moreover 
  the 
  method 
  of 
  taking 
  

   readings, 
  previously 
  described, 
  made 
  the 
  point 
  of 
  saturation, 
  

   when 
  there 
  was 
  much 
  fatigue, 
  less 
  marked. 
  Although 
  the 
  

   vacuum 
  was 
  good 
  whenever 
  readings 
  were 
  taken, 
  saturation, 
  in 
  

   the 
  case 
  of 
  the 
  scraped 
  metals, 
  was 
  not 
  always 
  complete 
  until 
  

   an 
  accelerating 
  potential 
  of 
  4 
  to 
  6 
  volts 
  was 
  applied. 
  This 
  

   lack 
  of 
  complete 
  saturation 
  may 
  have 
  been 
  due 
  to 
  reflected 
  

   electrons. 
  

  

  All 
  the 
  curves 
  show 
  that 
  the 
  point 
  of 
  saturation 
  is 
  inde- 
  

   pendent 
  of 
  the 
  wave 
  length 
  used. 
  This 
  is 
  a 
  confirmation 
  of 
  

   the 
  results 
  of 
  Kichardson 
  and 
  Compton, 
  and 
  is 
  seen 
  to 
  be 
  true 
  

   of 
  the 
  scraped 
  as 
  well 
  as 
  of 
  the 
  unscraped 
  metals. 
  In 
  the 
  

   Copper-Copper 
  curve 
  the 
  saturation 
  for 
  the 
  unscraped 
  metal 
  is 
  

   seen 
  to 
  occur 
  very 
  exactly 
  at 
  zero 
  volts. 
  In 
  the 
  Aluminium- 
  

   Aluminium 
  curve 
  this 
  point 
  comes 
  a 
  little 
  to 
  the 
  right 
  of 
  the 
  

   current 
  axis, 
  probably 
  due 
  to 
  the 
  fact 
  that 
  the 
  treatment 
  

  

  