﻿746 
  ProE. 
  0. 
  W. 
  Richardson 
  on 
  the 
  Specific 
  

  

  slit 
  between 
  cc, 
  so 
  as 
  to 
  allow 
  for 
  the 
  curvature 
  of 
  the 
  paths 
  

   of 
  the 
  ions. 
  The 
  whole 
  of 
  the 
  part 
  of 
  the 
  apparatus 
  com- 
  

   prised 
  by 
  V 
  1 
  1 
  V 
  was 
  surrounded 
  by 
  an 
  earth-connected 
  

   jacket 
  of 
  copper 
  foil, 
  not 
  shown, 
  so 
  as 
  to 
  avoid 
  creeping 
  and 
  

   induction 
  effects 
  as 
  far 
  as 
  possible. 
  This 
  jacket 
  was 
  of 
  course 
  

   provided 
  with 
  suitable 
  openings 
  for 
  the 
  electrometer 
  con- 
  

   nexions 
  k 
  and 
  /. 
  The 
  carriage 
  was 
  moved 
  by 
  a 
  rod 
  o, 
  

   provided 
  with 
  a 
  screw-thread 
  which 
  worked 
  in 
  a 
  fixed 
  nut 
  q 
  r 
  

   held 
  by 
  a 
  screw 
  in 
  the 
  outer 
  fixed 
  tube 
  p. 
  It 
  was 
  kept 
  

   pressed 
  against 
  the 
  end 
  of 
  the 
  rod 
  by 
  means 
  of 
  a 
  rubber 
  band 
  

   passing 
  over 
  the 
  upper 
  end 
  of 
  the 
  carriage 
  c, 
  and 
  the 
  lower 
  

   end 
  b 
  of 
  the 
  ebonite 
  frame. 
  The 
  rod 
  was 
  also 
  guided 
  by 
  

   fixed 
  rings 
  at 
  r 
  and 
  below 
  q. 
  The 
  lower 
  end 
  of 
  the 
  ebonite 
  

   frame 
  screwed 
  on 
  to 
  the 
  outer 
  tube 
  p 
  9 
  so 
  as 
  always 
  to 
  be 
  in 
  a 
  

   fixed 
  position 
  when 
  mounted. 
  The 
  rest 
  of 
  the 
  arrangements 
  

   will 
  be 
  best 
  understood 
  by 
  referring 
  to 
  fig. 
  2. 
  

  

  It 
  is 
  essential 
  that 
  the 
  foregoing 
  part 
  of 
  the 
  apparatus 
  

   should 
  be 
  in 
  a 
  very 
  good 
  vacuum. 
  This 
  was 
  secured 
  by 
  

   placing 
  it 
  in 
  the 
  tube 
  Q 
  : 
  which 
  fitted 
  on 
  the 
  ground-glass 
  

   stopper 
  s. 
  The 
  stopper 
  was 
  cemented 
  on 
  to 
  the 
  outside 
  of 
  

   the 
  brass 
  tube 
  p. 
  The 
  rod 
  o 
  could 
  be 
  turned 
  by 
  means 
  of 
  the 
  

   divided 
  head 
  T 
  from 
  outside 
  the 
  apparatus. 
  In 
  the 
  first 
  

   form 
  of 
  the 
  apparatus 
  the 
  tube 
  Q 
  x 
  was 
  inverted 
  and 
  an 
  air- 
  

   tight 
  joint 
  was 
  secured 
  by 
  simply 
  slipping 
  a 
  thick-walled 
  

   rubber 
  tube 
  over 
  the 
  tube 
  p 
  and 
  the 
  rod 
  o, 
  and 
  drowning 
  it 
  

   with 
  mercury. 
  But 
  this 
  was 
  found 
  to 
  be 
  unsatisfactory 
  

   owing 
  to 
  the 
  mercury 
  gradually 
  forcing 
  itself 
  into 
  the 
  

   apparatus 
  and 
  settling 
  in 
  places 
  where 
  it 
  was 
  not 
  wanted* 
  

   The 
  arrangement 
  shown 
  was 
  therefore 
  adopted 
  in 
  which 
  the 
  

   tube 
  Qj 
  and 
  adjacent 
  parts 
  rest 
  on 
  the 
  top 
  of 
  a 
  barometer 
  

   column. 
  This 
  was 
  contained 
  in 
  the 
  tube 
  Q 
  3 
  , 
  which 
  was 
  

   cemented 
  with 
  sealing-wax 
  to 
  the 
  brass 
  tube 
  p, 
  just 
  below 
  

   the 
  stopper 
  s. 
  A 
  side 
  tube 
  near 
  the 
  top 
  led 
  to 
  the 
  pump, 
  

   McLeod 
  gauge, 
  and 
  pentoxide 
  bulbs. 
  The 
  lower 
  end 
  of 
  the 
  

   glass 
  tube 
  was 
  drawn 
  down 
  so 
  that 
  the 
  rod 
  could 
  just 
  com- 
  

   fortably 
  turn 
  in 
  it, 
  and 
  the 
  thick 
  rubber 
  tube 
  R 
  was 
  slipped 
  

   over 
  the 
  joint. 
  The 
  mercury 
  entered 
  from 
  the 
  vessel 
  Q 
  2 
  by 
  

   means 
  of 
  the 
  side 
  tube. 
  The 
  level 
  of 
  the 
  mercury 
  at 
  Q 
  2 
  was 
  

   always 
  kept 
  above 
  R, 
  so 
  that 
  there 
  was 
  no 
  tendency 
  for 
  air 
  

   to 
  enter 
  at 
  the 
  joint. 
  An 
  earth-connected 
  wire, 
  not 
  shown, 
  

   dipped 
  permanently 
  into 
  Q 
  2 
  . 
  Another 
  wire 
  dipping 
  into 
  a 
  

   small 
  quantity 
  of 
  mercury 
  placed 
  in 
  the 
  stopper 
  s 
  ensured 
  

   that 
  the 
  copper 
  shield, 
  the 
  brass 
  plate 
  nn, 
  and 
  the 
  carriage 
  e 
  

   (fig. 
  1) 
  were 
  kept 
  at 
  zero 
  potential. 
  To 
  ensure 
  satisfactory 
  

   communication 
  between 
  Q 
  x 
  and 
  the 
  pump, 
  the 
  holes 
  uu' 
  were 
  

   bored 
  in 
  the 
  brass 
  tube 
  p. 
  The 
  micrometer 
  head 
  T 
  was 
  

  

  