﻿Radium 
  Emanation 
  in 
  the 
  Atmosphere. 
  587 
  

  

  with 
  the 
  calcium 
  chloride 
  cup 
  was 
  obtained 
  by 
  a 
  rod 
  which 
  

   passed 
  through 
  a 
  hole 
  in 
  the 
  side 
  of 
  the 
  shield-tube. 
  This 
  

   rod 
  was 
  connected 
  with 
  a 
  potentiometer-circuit 
  consisting 
  

   o£ 
  a 
  2-volt 
  cell 
  and 
  a 
  resistance-box; 
  by 
  these 
  means 
  the 
  

   cup, 
  and 
  therefore 
  one 
  pair 
  of 
  quadrants, 
  could 
  be 
  charged, 
  

   earthed, 
  or 
  insulated 
  at 
  will. 
  

  

  The 
  quadrant 
  electrometer 
  used 
  was 
  of 
  the 
  Dolezalek 
  

   pattern, 
  the 
  insulation 
  of 
  the 
  quadrants 
  and 
  terminals 
  being 
  

   amber 
  and 
  the 
  suspension-fibre 
  a 
  long 
  phosphor-bronze 
  strip. 
  

   The 
  case 
  of 
  the 
  electrometer 
  and 
  one 
  pair 
  of 
  quadrants 
  were 
  

   earthed. 
  The 
  needle 
  was 
  kept 
  at 
  a 
  potential 
  of 
  +80 
  volts 
  by 
  

   being 
  connected 
  to 
  a 
  battery 
  of 
  small 
  cells, 
  the 
  negative 
  ends 
  

   of 
  the 
  battery 
  being 
  earthed. 
  For 
  a 
  potential-difference 
  of 
  

   one 
  volt 
  between 
  the 
  quadrants 
  a 
  deflexion 
  of 
  90 
  cms. 
  on 
  a 
  

   scale 
  two 
  metres 
  distant 
  was 
  obtained. 
  

  

  The 
  testing 
  vessel 
  could 
  be 
  earthed 
  or 
  charged 
  by 
  connexion 
  

   to 
  a 
  battery 
  of 
  small 
  cells, 
  the 
  other 
  end 
  of 
  the 
  cells 
  being 
  

   earthed. 
  Usually 
  a 
  potential 
  of 
  —280 
  volts 
  was 
  applied 
  to 
  

   the 
  vessel. 
  This 
  was 
  sufficient 
  to 
  produce 
  saturation 
  for 
  all 
  

   the 
  leaks 
  that 
  I 
  had 
  occasion 
  to 
  measure. 
  Water 
  resistances 
  

   were 
  inserted 
  between 
  the 
  vessel 
  and 
  these 
  cells, 
  as 
  also 
  

   between 
  the 
  electrometer-needle 
  and 
  its 
  cells, 
  to 
  lessen 
  any 
  

   dangers 
  that 
  might 
  have 
  occurred 
  in 
  short-circuiting;. 
  

  

  During 
  the 
  experiments 
  it 
  was 
  found 
  that 
  if 
  the 
  air 
  in 
  the 
  

   testing 
  vessel 
  was 
  left 
  to 
  itself, 
  the 
  air 
  leak 
  gradually 
  increased 
  

   for 
  some 
  days; 
  but 
  that 
  fairly 
  steady 
  air 
  leaks 
  could 
  be 
  obtained 
  

   if 
  the 
  vessel 
  was 
  exhausted 
  every 
  day. 
  The 
  usual 
  value 
  of 
  

   the 
  air 
  leak 
  was 
  about 
  1*0 
  cm. 
  per 
  minute 
  on 
  the 
  scale 
  

   described 
  above. 
  This 
  is 
  equivalent 
  to 
  a 
  production 
  of 
  about 
  

   28 
  ions 
  per 
  c.c. 
  per 
  sec. 
  

  

  Sometimes 
  complications 
  were 
  introduced 
  by 
  the 
  calcium 
  

   chloride 
  in 
  the 
  drying-tube 
  next 
  the 
  testing 
  vessel. 
  If 
  the 
  

   vessel 
  was 
  exhausted 
  and 
  then 
  filled 
  quickly 
  through 
  the 
  

   drying-tube, 
  an 
  abnormally 
  large 
  leak 
  was 
  often 
  obtained 
  

   directly 
  afterwards, 
  although 
  great 
  care 
  was 
  taken 
  to 
  prevent 
  

   any 
  dust 
  entering 
  the 
  vessel 
  with 
  the 
  stream 
  of 
  air. 
  This 
  leak 
  

   gradually 
  decayed 
  to 
  the 
  value 
  of 
  the 
  normal 
  leak 
  in 
  about 
  

   8 
  minutes, 
  the 
  curve 
  of 
  decay 
  being 
  very 
  nearly 
  exponential 
  

   and 
  the 
  time 
  of 
  half 
  value 
  being 
  about 
  a 
  minute. 
  This 
  

   abnormal 
  leak 
  was, 
  however, 
  not 
  always 
  obtained; 
  it 
  seemed 
  

   to 
  depend 
  on 
  the 
  condition 
  of 
  the 
  calcium 
  chloride, 
  and 
  

   whenever 
  anything 
  peculiar 
  was 
  observed 
  in 
  the 
  ionization- 
  

   current 
  obtained 
  directly 
  after 
  filling 
  the 
  can, 
  a 
  check 
  expe- 
  

   riment 
  was 
  performed 
  as 
  soon 
  after 
  as 
  possible 
  by 
  taking 
  

   another 
  leak 
  after 
  a 
  quick 
  filling 
  of 
  the 
  vessel. 
  

  

  The 
  various 
  insulations 
  behaved 
  fairly 
  well 
  during 
  the 
  

  

  