﻿564 
  Mr. 
  K. 
  T. 
  Compton 
  and 
  Prof. 
  0. 
  W. 
  Richardson 
  on 
  

  

  Caesium. 
  — 
  The 
  results 
  in 
  the 
  case 
  of 
  csesium 
  are 
  illustrated 
  

   by 
  fig. 
  7, 
  curve 
  1. 
  The 
  energy 
  calibration 
  in 
  this 
  case 
  was 
  

   less 
  satisfactory 
  than 
  in 
  the 
  preceding 
  cases, 
  since 
  the 
  

   galvanometer 
  was 
  rather 
  unsteady. 
  On 
  the 
  other 
  hand, 
  the 
  

   photoelectric 
  measurements 
  were 
  exceedingly 
  satisfactory. 
  

  

  Fig-. 
  7. 
  

  

  70 
  

  

  60 
  

   50 
  

  

  40 
  

   30 
  

  

  N 
  

  

  Cs 
  

  

  

  

  

  

  

  -"Tid 
  

  

  

  

  

  

  

  

  

  

  

  

  

  *\ 
  

  

  

  

  

  

  •/ 
  

  

  / 
  ■ 
  

  

  

  

  

  

  V 
  

  

  

  

  

  4 
  

  

  

  

  

  

  

  

  

  

  / 
  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  / 
  

  

  

  

  

  

  

  

  

  

  

  y 
  

  

  

  

  

  

  

  

  

  

  

  50 
  60 
  10 
  80 
  90 
  

  

  100 
  110 
  

   V 
  (lO) 
  15 
  

  

  110 
  130 
  140 
  150 
  

  

  Evidently 
  the 
  method 
  employed 
  in 
  purifying 
  and 
  distilling 
  

   the 
  csesium 
  made 
  possible 
  a 
  better 
  vacuum, 
  and 
  consequently 
  

   there 
  was 
  less 
  fatigue 
  than 
  in 
  the 
  preceding 
  cases. 
  

  

  Discussion 
  of 
  Results. 
  

  

  The 
  most 
  striking 
  conclusion 
  to 
  be 
  drawn 
  from 
  these 
  

   results 
  is 
  that 
  there 
  are 
  two 
  maxima 
  in 
  the 
  sensitiveness- 
  

   frequency 
  curve, 
  instead 
  of 
  but 
  one 
  maximum, 
  as 
  indicated 
  

   bv 
  the 
  theory 
  and 
  by 
  the 
  experiments 
  of 
  Pohl 
  and 
  Pringsheim. 
  

   In 
  the 
  Pohl 
  and 
  Pringsheim 
  curve 
  in 
  fig. 
  1 
  we 
  see 
  that 
  the 
  

   " 
  normal 
  " 
  photoelectric 
  effect, 
  represented 
  by 
  abed, 
  reaches 
  

   a 
  maximum 
  similar 
  to 
  that 
  reached 
  by 
  the 
  " 
  selective 
  " 
  effect 
  

   b 
  c 
  d, 
  but 
  farther 
  out 
  in 
  the 
  ultra-violet. 
  Experiments 
  with 
  

   Rontgen 
  rays 
  indicate 
  that 
  the 
  " 
  normal 
  " 
  effect 
  does 
  not 
  

   increase 
  indefinitely 
  with 
  the 
  frequency. 
  The 
  fact 
  that 
  we 
  

   were 
  able 
  to 
  discover 
  this 
  second 
  maximum 
  was 
  due 
  to 
  the 
  use 
  

   of 
  very 
  electropositive 
  metals 
  and 
  to 
  the 
  sensitiveness 
  of 
  our 
  

   energy 
  calibration 
  system, 
  which 
  enabled 
  the 
  investigation 
  

   to 
  be 
  carried 
  further 
  into 
  the 
  ultra-violet 
  than 
  ever 
  before. 
  

  

  