﻿588 
  Mr. 
  A. 
  B. 
  Wood 
  on 
  the 
  Range 
  of 
  the 
  

  

  others 
  to 
  prove 
  the 
  existence 
  of 
  two, 
  while 
  again 
  others 
  

   indicated 
  no 
  definite 
  critical 
  pressure 
  at 
  all. 
  It 
  thus 
  became 
  

   necessary 
  to 
  find 
  out 
  the 
  causes 
  of 
  these 
  inconsistencies. 
  One 
  

   source 
  of 
  trouble 
  was 
  traced 
  to 
  the 
  presence 
  of 
  water-vapour 
  

   in 
  the 
  apparatus. 
  The 
  water-vapour, 
  when 
  suspended 
  in 
  the 
  

   air, 
  might 
  possibly 
  condense 
  to 
  some 
  extent 
  on 
  the 
  active 
  

   deposit 
  and 
  on 
  the 
  recoiling 
  atoms, 
  and 
  thus 
  alter 
  the 
  con- 
  

   ditions 
  of 
  recoil 
  considerably. 
  Again, 
  it 
  was 
  discovered 
  that 
  

   the 
  amount 
  of 
  dust 
  admitted 
  into 
  the 
  apparatus 
  had 
  also 
  a 
  

   striking 
  effect 
  on 
  the 
  activity 
  received 
  by 
  the 
  hemisphere. 
  

   This 
  has 
  also 
  been 
  observed 
  by 
  Mr. 
  Walmsley 
  * 
  in 
  some 
  

   experiments 
  with 
  actinium 
  emanation 
  and 
  active 
  deposit. 
  

   With 
  a 
  view 
  to 
  finding 
  the 
  cause 
  of 
  this, 
  we 
  performed 
  a 
  few 
  

   simple 
  experiments 
  together, 
  and 
  succeeded 
  in 
  showing 
  that 
  

   ordinary 
  atmospheric 
  dust 
  in 
  the 
  laboratory 
  carried 
  a 
  negative 
  

   charge. 
  Hence 
  any 
  recoiling 
  atoms 
  which 
  deposit 
  on 
  a 
  

   charged 
  dust- 
  particle 
  will 
  be 
  carried 
  to 
  the 
  positive 
  electrode. 
  

   In 
  order 
  to 
  avoid 
  these 
  troubles 
  all 
  gas 
  admitted 
  into 
  the 
  

   apparatus 
  was 
  carefully 
  dried 
  over 
  P 
  2 
  5 
  and 
  filtered 
  through 
  

   tightly-packed 
  cotton-wool. 
  

  

  In 
  performing 
  an 
  experiment, 
  the 
  source 
  a 
  was 
  placed 
  in 
  

   position 
  and 
  a 
  tinfoil 
  hemisphere 
  exposed 
  to 
  it 
  for 
  15 
  minutes 
  

   at 
  the 
  desired 
  pressure. 
  Air 
  which 
  had 
  been 
  previously 
  dried 
  

   and 
  filtered 
  was 
  then 
  admitted 
  into 
  the 
  apparatus 
  and 
  the 
  

   /3-ray 
  activity 
  of 
  the 
  hemisphere 
  was 
  measured. 
  The 
  observed 
  

   activities 
  were 
  corrected 
  (1) 
  for 
  the 
  decay 
  of 
  the 
  source 
  a 
  

   (period 
  10*6 
  hours) 
  during 
  the 
  experiment, 
  and 
  (2) 
  for 
  the 
  

   decay 
  of 
  the 
  thorium 
  D 
  (period 
  3'lmins.) 
  on 
  the 
  hemisphere 
  

   daring 
  the 
  time 
  of 
  its 
  removal 
  from 
  the 
  bell-jar 
  and 
  the 
  

   measurement 
  of 
  its 
  activity. 
  

  

  The 
  results 
  of 
  a 
  typical 
  set 
  of 
  measurements 
  with 
  the 
  recoil 
  

   from 
  thorium 
  active 
  deposit 
  in 
  air 
  are 
  shown 
  in 
  fig. 
  2. 
  It 
  

   will 
  be 
  seen 
  at 
  once 
  that 
  the 
  recoiling 
  beam 
  is 
  not 
  homo- 
  

   geneous, 
  as 
  one 
  would 
  have 
  expected 
  from 
  the 
  experimental 
  

   arrangement. 
  No 
  " 
  critical 
  pressure 
  " 
  is 
  observed 
  ; 
  the 
  

   activity 
  at 
  first 
  falls 
  off 
  rapidly 
  with 
  increase 
  of 
  pressure 
  

   and 
  afterwards 
  less 
  rapidly, 
  eventually 
  becoming 
  nearly 
  

   constant 
  with 
  further 
  increase 
  of 
  pressure. 
  The 
  activities 
  

   observed 
  at 
  pressures 
  above 
  4 
  mm. 
  are 
  in 
  all 
  probability 
  due 
  

   to 
  the 
  diffusion 
  of 
  uncharged 
  recoil 
  atoms 
  to 
  the 
  hemisphere. 
  

   It 
  is 
  well 
  known 
  that 
  the 
  positively 
  charged 
  recoil 
  atoms 
  

   recombine 
  with 
  negative 
  ions 
  and 
  become 
  neutral. 
  Since 
  

   there 
  was 
  a 
  comparatively 
  intense 
  ionization 
  in 
  the 
  vessel, 
  a 
  

   small 
  portion 
  of 
  the 
  recoil 
  atoms 
  would 
  lose 
  their 
  charge. 
  

   These 
  would 
  diffuse 
  to 
  the 
  boundaries 
  and 
  be 
  unaffected 
  by 
  

   * 
  Phil. 
  Mag. 
  Sept, 
  1913. 
  

  

  