﻿Helium 
  from 
  the 
  Primary 
  Radio-Elements. 
  521 
  

  

  in 
  progress, 
  many 
  changes 
  and 
  improvements 
  were 
  introduced. 
  

   So 
  far 
  as 
  the 
  flask 
  and 
  its 
  attached 
  mercury 
  tap 
  is 
  concerned, 
  

   the 
  design 
  then 
  adopted 
  has 
  been 
  adhered 
  to 
  with 
  but 
  slight 
  

   modification 
  in 
  all 
  the 
  later 
  experiments, 
  and 
  has 
  been 
  found 
  

   to 
  work 
  perfectly, 
  and 
  will 
  therefore 
  be 
  described 
  in 
  detail. 
  

   But 
  frequent 
  changes 
  have 
  been 
  made 
  from 
  time 
  to 
  time 
  in 
  

   the 
  method 
  of 
  performing 
  the 
  testing 
  operations. 
  As 
  finality 
  

   in 
  this 
  direction 
  has 
  perhaps 
  not 
  even 
  yet 
  been 
  reached, 
  the 
  

   detailed 
  description 
  of 
  this 
  part 
  of 
  the 
  work 
  will 
  be 
  deferred. 
  

   The 
  apparatus 
  in 
  its 
  present 
  form 
  is 
  shown 
  in 
  the 
  diagram 
  in 
  

   fig. 
  1 
  (p. 
  522), 
  and 
  the 
  photographs 
  of 
  the 
  actual 
  thorium 
  

   apparatus, 
  from 
  the 
  front 
  and 
  from 
  the 
  sides, 
  are 
  shown 
  

   in 
  Plate 
  XX. 
  The 
  earlier 
  failures 
  had 
  shown 
  the 
  difficulty, 
  

   first 
  of 
  freeing 
  a 
  large 
  quantity 
  of 
  liquid 
  perfectly 
  from 
  

   dissolved 
  air, 
  and 
  secondly 
  of 
  getting 
  it 
  under 
  these 
  circum- 
  

   stances 
  to 
  boil 
  regularly. 
  The 
  dangerous 
  percussive 
  boiling, 
  

   which 
  made 
  the 
  earlier 
  experiments 
  so 
  precarious, 
  was 
  re- 
  

   duced, 
  and 
  at 
  the 
  same 
  time 
  the 
  dissolved 
  air 
  was 
  efficiently 
  

   and 
  rapidly 
  removed 
  by 
  sealing 
  into 
  the 
  flask 
  (B) 
  a 
  tube 
  

   reaching 
  to 
  the 
  bottom 
  and 
  connected 
  to 
  a 
  voltameter 
  (A) 
  for 
  

   generating 
  a 
  mixture 
  of 
  hydrogen 
  and 
  oxygen. 
  An 
  all-glass 
  

   vertical 
  condenser 
  (0) 
  with 
  double 
  circulation 
  of 
  water 
  was 
  

   sealed 
  to 
  the 
  neck 
  of 
  the 
  flask, 
  and 
  provided 
  at 
  the 
  upper 
  end 
  

   with 
  a 
  large 
  bulb. 
  This 
  was 
  connected 
  to 
  a 
  trap 
  (D) 
  formed 
  

   by 
  a 
  single 
  drop 
  of 
  mercury, 
  for 
  the 
  purpose 
  of 
  preventing 
  

   the 
  expelled 
  gas 
  from 
  diffusing 
  back 
  into 
  the 
  flask. 
  During 
  

   the 
  preliminary 
  removal 
  of 
  the 
  air 
  the 
  by-pass 
  (a?) 
  was 
  open, 
  

   and 
  then 
  it 
  was 
  sealed 
  as 
  shown. 
  E 
  is 
  a 
  glass 
  chamber 
  for 
  

   combining 
  the 
  hydrogen 
  and 
  oxygen 
  by 
  means 
  of 
  a 
  discharge 
  

   from 
  a 
  coil, 
  and 
  for 
  condensing 
  the 
  bulk 
  of 
  the 
  water 
  produced. 
  

   F 
  is 
  the 
  mercury 
  tap. 
  Gr 
  is 
  the 
  U-tube 
  to 
  condense 
  water- 
  

   vapour, 
  which 
  in 
  the 
  later 
  experiments 
  has 
  been 
  immersed 
  in 
  

   liquid 
  air. 
  H 
  is 
  the 
  calcium 
  furnace, 
  which, 
  after 
  the 
  expe- 
  

   riment, 
  is 
  sealed 
  off 
  and 
  filled 
  with 
  mercury 
  from 
  the 
  

   crucible 
  (L). 
  K 
  is 
  the 
  spectrum-tube 
  described 
  later. 
  J 
  is 
  

   a 
  subsidiary 
  volume, 
  only 
  used 
  in 
  the 
  more 
  recent 
  tests, 
  the 
  

   purpose 
  of 
  which 
  is 
  to 
  neutralize 
  the 
  effect 
  of 
  the 
  volume 
  of 
  

   the 
  U-tube. 
  The 
  device 
  is 
  copied 
  from 
  that 
  employed 
  by 
  

   Ramsay 
  & 
  Cameron 
  ( 
  Journ. 
  Chem. 
  Soc. 
  Trans. 
  1907, 
  p. 
  1598, 
  

   fig. 
  1). 
  The 
  U-tube 
  G 
  must 
  necessarily 
  have 
  a 
  somewhat 
  

   large 
  volume 
  to 
  prevent 
  its 
  becoming 
  choked 
  with 
  ice. 
  At 
  

   the 
  temperature 
  of 
  liquid 
  air 
  its 
  volume 
  is 
  virtually 
  trebled, 
  

   whereas 
  the 
  volume 
  of 
  the 
  calcium 
  furnace 
  at 
  the 
  high 
  tem- 
  

   perature 
  is 
  virtually 
  halved. 
  Hence 
  an 
  inconveniently 
  large 
  

   fraction 
  of 
  the 
  gas 
  remains 
  in 
  the 
  U-tube 
  when 
  it 
  is 
  sealed 
  off. 
  

   To 
  avoid 
  this, 
  J 
  is 
  surrounded 
  with 
  a 
  cup 
  of 
  moist 
  filter-paper 
  

  

  