﻿in 
  Their 
  Thermodynamic 
  Relations. 
  

  

  163 
  

  

  and 
  2*5 
  cm 
  in 
  diameter, 
  with 
  walls 
  *6 
  cm 
  thick. 
  The 
  interior 
  is 
  

   threaded 
  (12 
  to 
  inch) 
  at 
  the 
  ends, 
  and 
  screw 
  pings 
  are 
  provided 
  

   fitting 
  the 
  threads 
  snugly 
  {CD). 
  

  

  In 
  order 
  to 
  secure 
  a 
  gas-tight 
  joint 
  for 
  temperatures 
  below 
  

   150°, 
  each 
  of 
  the 
  pings 
  is 
  covered 
  with 
  a 
  thin 
  layer 
  of 
  solder 
  

   in 
  the 
  usual 
  way, 
  by 
  dipping 
  it 
  first 
  in 
  soldering 
  salts, 
  and 
  then 
  

   in 
  melted 
  solder. 
  For 
  temperatures 
  below 
  250°, 
  a 
  layer 
  of 
  

   lead 
  is 
  applied 
  in 
  the 
  same 
  way. 
  For 
  temperatures 
  above 
  this 
  

   a 
  layer 
  of 
  zinc, 
  and 
  at 
  red 
  heat 
  a 
  layer 
  of 
  copper 
  answers 
  less 
  

   satisfactorily. 
  The 
  copper 
  coating 
  is 
  put 
  on 
  by 
  dipping 
  the 
  

   hot 
  plug 
  in 
  borax 
  and 
  then 
  in 
  a 
  crucible 
  of 
  melted 
  copper. 
  

  

  A 
  jf 
  

  

  hc<^\\\\\\\\\\\\\\\\\\j 
  

  

  \\\\\\\\\\\ 
  \\w\\\\\v\ 
  v\^ 
  

   B 
  

  

  Unfortunately 
  the 
  copper 
  is 
  hardened 
  by 
  this 
  process 
  and 
  I 
  

   have 
  only 
  been 
  able 
  to 
  use 
  the 
  screw 
  for 
  small 
  diameters. 
  

   Solder 
  adheres 
  most 
  perfectly 
  and 
  less 
  stress 
  is 
  needed 
  in 
  forc- 
  

   ing 
  the 
  screw 
  plug 
  home. 
  Lead, 
  in 
  spite 
  of 
  its 
  greater 
  softness, 
  

   offers 
  much 
  more 
  viscous 
  resistance 
  during 
  this 
  operation 
  ; 
  a 
  

   noteworthy 
  result. 
  Tubes 
  continually 
  used 
  in 
  this 
  way 
  are 
  at 
  

   length 
  liable 
  to 
  split 
  longitudinally, 
  but 
  after 
  several 
  hundred 
  

   experiments 
  per 
  tube 
  I 
  have 
  only 
  had 
  one 
  accident 
  of 
  this 
  

   kind. 
  The 
  thread 
  is 
  apt 
  to 
  suffer 
  serious 
  wear 
  and 
  tear 
  before 
  

   the 
  tube 
  gives 
  out. 
  The 
  joint 
  obtained 
  as 
  above 
  is 
  apparently 
  

   quite 
  gas-tight, 
  and 
  I 
  have 
  frequently 
  heated 
  carbon 
  disulphide 
  

   above 
  200° 
  for 
  hours, 
  without 
  appreciable 
  loss. 
  Nor 
  is 
  it 
  nec- 
  

   essary 
  to 
  use 
  all 
  the 
  threads 
  at 
  once 
  ; 
  a 
  fresh 
  insertion 
  of 
  

   two 
  complete 
  threads 
  usually 
  suffices, 
  so 
  that 
  the 
  screw 
  bolt 
  

   lasts 
  throughout 
  several 
  experiments 
  before 
  requiring 
  to 
  be 
  

   recovered. 
  Pressures 
  above 
  1000 
  atm. 
  'were 
  applied 
  without 
  

   leakage. 
  

  

  To 
  heat 
  the 
  tube 
  to 
  a 
  given 
  constant 
  temperature, 
  vapor 
  

   baths 
  of 
  the 
  form 
  fig. 
  2 
  are 
  available. 
  This 
  consists 
  of 
  a 
  

   cylinder 
  AT, 
  10 
  cm 
  in 
  diameter, 
  T0 
  cm 
  long, 
  closed 
  at 
  the 
  ends, 
  

   except 
  for 
  the 
  axially 
  perforating 
  eccentric 
  cylinder 
  GG. 
  The 
  

   latter 
  is 
  large 
  enough 
  to 
  receive 
  the 
  retort 
  AJB, 
  snugly, 
  and 
  

   asbestos 
  plugs, 
  Jill, 
  guard 
  against 
  loss 
  of 
  heat 
  at 
  the 
  ends. 
  

   The 
  vapor 
  bath 
  is 
  made 
  of 
  sheet 
  iron 
  brazed 
  together. 
  The 
  

   liquid 
  to 
  be 
  boiled, 
  k, 
  is 
  charged 
  into 
  the 
  bottom 
  of 
  the 
  outer 
  

   cylinder, 
  and 
  vapors 
  escape 
  by 
  a 
  lateral 
  vent 
  of 
  gas 
  pipe 
  

  

  