﻿Thermal 
  Conductivity 
  of 
  Glass. 
  

  

  601 
  

  

  K 
  

  

  Fig. 
  2. 
  

   E 
  

  

  M 
  

  

  DE 
  was 
  fused 
  into 
  AA 
  2 
  at 
  D, 
  and 
  an 
  insulated 
  thermo- 
  

   junction 
  was 
  put 
  down 
  into 
  the 
  mercury 
  in 
  D. 
  The 
  tube 
  

   DE 
  (fig. 
  2) 
  was 
  surrounded 
  by 
  a 
  vacuum 
  

   jacket 
  HKMNL. 
  It 
  was 
  hoped 
  that 
  the 
  

   temperature 
  of 
  the 
  mercury 
  in 
  D 
  would 
  

   ultimately 
  be 
  the 
  same 
  as 
  that 
  of 
  the 
  mercury 
  

   in 
  the 
  capillary. 
  This 
  arrangement 
  gave 
  

   very 
  unsatisfactory 
  results, 
  and 
  the 
  readings 
  

   taken 
  indicated 
  clearly 
  that 
  the 
  mercury 
  in 
  

   the 
  small 
  bulb 
  D 
  was 
  at 
  a 
  higher 
  tempe- 
  

   rature 
  than 
  that 
  in 
  the 
  capillary 
  tube. 
  

  

  § 
  5. 
  The 
  arrangement 
  finally 
  adopted 
  is 
  

   shown 
  in 
  fig. 
  3. 
  A 
  hole 
  was 
  drilled 
  into 
  the 
  

   capillary 
  at 
  D, 
  and 
  an 
  insulated 
  thermo- 
  

   junction 
  was 
  inserted 
  through 
  this 
  hole 
  

   directly 
  into 
  the 
  mercury 
  in 
  the 
  capillary. 
  

   There 
  was 
  no 
  bulb 
  at 
  D, 
  and 
  consequently 
  

   the 
  cross-section 
  of 
  the 
  mercury 
  column 
  was 
  

   practically 
  undisturbed. 
  

  

  The 
  thermocouple 
  used 
  was 
  made 
  of 
  

   copper-constantan. 
  The 
  two 
  wires 
  pro- 
  

   ceeding 
  from 
  each 
  junction 
  were 
  carefully 
  insulated 
  from 
  

   one 
  another 
  by 
  means 
  of 
  rubber 
  tubes. 
  Each 
  pair 
  of 
  wires 
  

   fitted 
  into 
  a 
  narrow 
  glass 
  tube 
  about 
  eight 
  inches 
  

  

  was 
  

  

  Fig. 
  3. 
  

  

  E 
  

  

  XV.'A 
  VZJts^'^h 
  VZZZZZZZZSSYZZZ 
  

  

  VT7777777777T/. 
  . 
  777 
  77,'///. 
  , 
  J/»— 
  

  

  A 
  

  

  B 
  

  

  B, 
  

  

  long 
  and 
  open 
  at 
  both 
  ends. 
  The 
  junctions 
  themselves 
  

   protruded 
  about 
  five 
  millimetres 
  from 
  these 
  tubes 
  and 
  were 
  

   insulated 
  by 
  means 
  of 
  a 
  layer 
  of 
  varnish. 
  The 
  open 
  ends 
  of 
  

   the 
  glass 
  tubes, 
  next 
  to 
  the 
  junction, 
  were 
  then 
  closed 
  up 
  

   with 
  cement. 
  One 
  junction 
  was 
  inserted 
  into 
  the 
  mercury 
  

   column 
  through 
  the 
  small 
  bole 
  D 
  (fig. 
  3)', 
  and 
  the 
  glass 
  tube 
  

   DE 
  surrounding 
  the 
  junction 
  was 
  cemented 
  to 
  the 
  capillary 
  

  

  