﻿530 
  Mr. 
  A. 
  Fleck 
  on 
  the 
  

  

  seven 
  minutes 
  the 
  activity 
  of 
  the 
  plate 
  was 
  only 
  from 
  60 
  to 
  

   80 
  per 
  cent, 
  of 
  the 
  initial 
  activity, 
  and 
  this 
  activity 
  remained 
  

   constant 
  over 
  a 
  number 
  of 
  hours. 
  By 
  drawing 
  the 
  difference 
  

   curve 
  obtained 
  by 
  subtracting 
  this 
  constant 
  activity 
  it 
  was 
  

   seen 
  to 
  be 
  approximately 
  exponential, 
  and 
  that 
  the 
  radiation 
  

   decayed 
  to 
  half-value 
  in 
  from 
  one 
  to 
  one 
  and 
  a 
  half 
  minutes. 
  

   Fajans 
  and 
  G-ohring's 
  experiment 
  has 
  therefore 
  been 
  repeated 
  

   and 
  confirmed. 
  

  

  If 
  lead 
  chloride 
  is 
  precipitated 
  with 
  hydrochloric 
  acid 
  in 
  

   a 
  uranium 
  X 
  solution, 
  the 
  precipitate 
  collected 
  and 
  its 
  

   activity 
  measured, 
  then 
  this 
  rapid 
  initial 
  decay 
  is 
  again 
  

   observed. 
  The 
  proportional 
  decay 
  is 
  usually 
  about 
  the 
  same 
  as 
  

   in 
  the 
  case 
  of 
  the 
  lead 
  plate, 
  but 
  sometimes 
  amounts 
  to 
  about 
  

   50 
  per 
  cent, 
  of 
  the 
  initial 
  activity. 
  In 
  both 
  of 
  these 
  experi- 
  

   ments 
  the 
  quantity 
  of 
  activity 
  precipitated 
  with 
  the 
  lead 
  is 
  

   small 
  compared 
  with 
  the 
  total 
  quantity 
  present. 
  

  

  That 
  this 
  rapid 
  decay 
  is 
  not 
  due 
  to 
  diffusion 
  or 
  other 
  

   physical 
  cause 
  connected 
  with 
  the 
  lead 
  was 
  shown 
  in 
  twa 
  

   ways. 
  A 
  tray 
  of 
  thin 
  lead-foil 
  was 
  made, 
  into 
  which 
  a 
  

   weakly 
  acid 
  solution 
  of 
  uranium 
  X 
  was 
  placed, 
  and 
  after 
  

   five 
  minutes 
  the 
  solution 
  was 
  removed. 
  The 
  tray, 
  washed 
  

   and 
  dried 
  in 
  the 
  usual 
  way 
  with 
  alcohol, 
  was 
  then 
  placed 
  

   under 
  the 
  electroscope 
  in 
  an 
  inverted 
  position. 
  If 
  the 
  effect 
  

   previously 
  observed 
  had 
  been 
  due 
  to 
  diffusion, 
  then, 
  when 
  the 
  

   ft 
  activity 
  of 
  the 
  tray 
  was 
  measured 
  in 
  an 
  inverted 
  position, 
  

   an 
  initial 
  rise 
  should 
  have 
  been 
  observed. 
  No 
  such 
  rise 
  was 
  

   seen, 
  but, 
  on 
  the 
  contrary, 
  the 
  same 
  rapid 
  initial 
  decay 
  was 
  

   observed. 
  Radium 
  E 
  gives 
  off 
  j3 
  rays 
  not 
  greatly 
  differing 
  

   from 
  those 
  of 
  uranium 
  X, 
  and 
  if 
  the 
  effect 
  had 
  been 
  due 
  to 
  

   any 
  action 
  of 
  the 
  rays 
  on 
  lead, 
  it 
  would 
  have 
  been 
  expected 
  

   that 
  it 
  would 
  have 
  been 
  obtained 
  on 
  precipitating 
  lead 
  chloride- 
  

   in 
  a 
  strong 
  solution 
  of 
  radium 
  E. 
  When 
  this 
  was 
  done 
  

   no 
  rapid 
  initial 
  decay 
  was 
  observed 
  in 
  the 
  /3 
  activity 
  of 
  the 
  

   precipitate. 
  

  

  Experiments 
  were 
  made 
  by 
  precipitating 
  bismuth 
  oxy- 
  

   chloride 
  in 
  a 
  uranium 
  X 
  solution, 
  and 
  the 
  curve 
  obtained 
  by 
  

   measuring 
  the 
  j3 
  activity 
  of 
  one 
  such 
  precipitate 
  is 
  shown 
  in 
  

   fig. 
  1, 
  along 
  with 
  the 
  logarithmic 
  curve 
  obtained 
  by 
  sub- 
  

   tracting 
  the 
  final 
  activity 
  reached 
  after 
  nine 
  minutes 
  from 
  

   the 
  activity 
  at 
  any 
  time. 
  It 
  is 
  seen 
  that 
  the 
  activity 
  of 
  

   52 
  divs. 
  per 
  min., 
  measured 
  two 
  minutes 
  after 
  precipitation, 
  

   decays 
  to 
  an 
  activity 
  of 
  6*5 
  divs. 
  per 
  min., 
  showing 
  that 
  the 
  

   substance 
  is 
  in 
  a 
  relatively 
  pure 
  condition. 
  The 
  period 
  of 
  

   average 
  life 
  calculated 
  from 
  three 
  values 
  of 
  this 
  curve 
  is 
  1*44, 
  

   1*83, 
  and 
  1*63 
  minutes 
  — 
  an 
  average 
  value 
  of 
  1*627 
  minutes. 
  

   By 
  extrapolating 
  this 
  curve 
  backwards 
  to 
  the 
  instant 
  of 
  

  

  