﻿Helium 
  from 
  the 
  Primary 
  Radio-Elements. 
  519 
  

  

  acid 
  and 
  phosphorus 
  pentoxide 
  respectively, 
  was 
  erected 
  at 
  

   the 
  commencement 
  o£ 
  the 
  work, 
  and 
  has 
  been 
  kept 
  ex- 
  

   clusively 
  for 
  these 
  experiments. 
  In 
  making 
  a 
  test 
  for 
  helium, 
  

   the 
  calcium 
  furnace 
  is 
  first 
  pumped 
  out 
  as 
  completely 
  as 
  

   possible, 
  the 
  calcium 
  is 
  then 
  heated, 
  and 
  the 
  pumping 
  con- 
  

   tinued, 
  so 
  that 
  the 
  expelled 
  gas 
  serves 
  to 
  wash 
  out 
  the 
  last 
  

   traces 
  of 
  air. 
  The 
  pump 
  is 
  then 
  sealed 
  off, 
  and 
  communica- 
  

   tion 
  made 
  with 
  the 
  flask 
  through 
  the 
  cooled 
  U-tube, 
  and 
  

   the 
  solution 
  boiled. 
  The 
  calcium 
  furnace 
  is 
  sealed 
  off 
  and, 
  

   when 
  cold, 
  filled 
  in 
  the 
  dark 
  room 
  with 
  dry 
  mercury. 
  As 
  

   soon 
  as 
  the 
  spectrum-tube 
  conducts 
  at 
  all, 
  the 
  spectrum 
  is 
  

   rapidly 
  examined 
  with 
  a 
  Ramsay 
  train 
  of 
  prisms, 
  and 
  if 
  D 
  3 
  , 
  

   the 
  characteristic 
  line 
  of 
  the 
  helium, 
  is 
  observed, 
  the 
  tube 
  is 
  

   generally 
  sealed 
  off 
  and 
  examined 
  with 
  a 
  good 
  spectroscope. 
  

  

  The 
  first 
  and 
  also 
  the 
  second 
  series 
  of 
  experiments 
  with 
  

   thorium 
  and 
  uranium 
  came 
  to 
  a 
  premature 
  end 
  on 
  account 
  s\ 
  

   of 
  accidents; 
  but 
  the 
  third 
  series, 
  of 
  which 
  the 
  experiment 
  

   with 
  thorium 
  was 
  started 
  first, 
  has 
  been 
  more 
  successful, 
  and 
  ^^/{/XjCCkj 
  

   is 
  still 
  in 
  progress. 
  The 
  main 
  experimental 
  difficulty 
  at 
  first 
  

   encountered 
  was 
  air. 
  The 
  argon 
  in 
  a 
  few 
  cubic 
  centimetres 
  "~~ 
  

   of 
  air 
  is 
  sufficient 
  to 
  prevent 
  completely 
  the 
  detection 
  of 
  the 
  

   minute 
  amounts 
  of 
  helium 
  looked 
  for 
  in 
  these 
  experiments. 
  ■. 
  ' 
  

   To 
  free 
  a 
  litre 
  or 
  more 
  of 
  solution 
  from 
  every 
  trace 
  of 
  air, 
  

   and 
  to 
  preserve 
  it 
  for 
  months 
  or 
  years 
  in 
  this 
  condition 
  while 
  

   making 
  provision 
  for 
  the 
  complete 
  removal 
  of 
  the 
  gas 
  for 
  

   testing 
  at 
  will, 
  calls 
  for 
  very 
  perfect 
  arrangements, 
  and 
  the 
  

   method 
  employed 
  in 
  the 
  last 
  series 
  is 
  the 
  outcome 
  of 
  many 
  

   previous 
  failures. 
  In 
  the 
  first 
  series 
  of 
  experiments, 
  started 
  

   in 
  August 
  1905, 
  about 
  a 
  kilogram 
  each 
  of 
  uranium 
  nitrate 
  

   and 
  thorium 
  nitrate, 
  of 
  ordinary 
  commercial 
  quality, 
  were 
  

   employed. 
  The 
  exit 
  tube 
  from 
  the 
  flasks 
  was 
  joined 
  to 
  

   a 
  Y-tube, 
  the 
  vertical 
  limb 
  of 
  which 
  was 
  a 
  barometer 
  dipping 
  

   into 
  a 
  mercury 
  reservoir 
  which 
  could 
  be 
  raised 
  or 
  lowered, 
  

   sealing 
  or 
  opening 
  the 
  passage 
  through 
  the 
  upper 
  branches 
  

   of 
  the 
  Y-tube. 
  The 
  other 
  branch 
  of 
  the 
  Y 
  was 
  provided 
  

   with 
  a 
  tap, 
  and 
  the 
  pump, 
  &c. 
  could 
  thus 
  be 
  sealed 
  to 
  it 
  

   without 
  air 
  entering 
  the 
  solution. 
  In 
  this 
  way 
  the 
  air-tight- 
  

   ness 
  of 
  the 
  tap 
  was 
  depended 
  upon 
  only 
  during 
  the 
  actual 
  

   removal 
  of 
  the 
  gases 
  to 
  the 
  testing-furnace, 
  and 
  any 
  air 
  

   leaking 
  through 
  during 
  the 
  long 
  periods 
  of 
  accumulation 
  could 
  

   be 
  removed 
  before 
  lowering 
  the 
  mercury 
  seal. 
  This 
  worked 
  

   well 
  until 
  one 
  night 
  the 
  lubricant 
  of 
  the 
  tap 
  failed, 
  and 
  the 
  next 
  

   morning 
  air 
  was 
  found 
  to 
  be 
  bubbling 
  through 
  the 
  mercury 
  

   seal 
  into 
  the 
  flasks. 
  Before 
  this 
  ten 
  tests 
  had 
  been 
  made 
  — 
  

   six 
  with 
  the 
  uranium 
  and 
  four 
  with 
  the 
  thorium. 
  The 
  whole 
  

   of 
  six 
  former 
  and 
  the 
  first 
  three 
  of 
  the 
  latter 
  were 
  failures 
  

  

  