﻿Dr. 
  0. 
  Chree's 
  Notes 
  on 
  Tliermometry, 
  323 
  

  

  International 
  is 
  used 
  in 
  a 
  carefully 
  prescribed 
  way, 
  one 
  can 
  

   deduce 
  the 
  corresponding 
  temperature 
  on 
  the 
  scale 
  of 
  the 
  

   hydro 
  gen 
  thermometer 
  (at 
  least 
  throughout 
  the 
  range 
  — 
  20° 
  C. 
  

   to 
  100° 
  C.) 
  to 
  a 
  very 
  high 
  degree 
  of 
  accuracy. 
  In 
  a 
  physical 
  

   laboratory 
  the 
  probable 
  error 
  may 
  be 
  as 
  small 
  as 
  +0°002 
  0., 
  

   or 
  possibly 
  even 
  +0°001 
  C. 
  It 
  may 
  also 
  be 
  urged 
  that 
  the 
  

   reduction 
  of 
  the 
  thermometer 
  readings 
  by 
  means 
  of 
  the 
  tables 
  

   supplied 
  by 
  the 
  Bureau 
  International 
  requires 
  no 
  higher 
  

   arithmetical 
  powers 
  than 
  every 
  well 
  educated 
  man 
  should 
  

   possess. 
  

  

  On 
  the 
  other 
  hand, 
  it 
  must 
  be 
  conceded 
  that 
  the 
  existence 
  

   of 
  lengthy 
  tables 
  is 
  an 
  evil, 
  however 
  necessary, 
  both 
  on 
  account 
  

   of 
  the 
  very 
  appreciable 
  labour 
  their 
  application 
  entails, 
  and 
  

   on 
  account 
  of 
  the 
  large 
  increase 
  their 
  calculation 
  makes 
  to 
  

   the 
  prime 
  cost 
  of 
  the 
  instrument. 
  The 
  fact 
  that 
  for 
  high 
  

   accuracy 
  an 
  ice 
  reading 
  is 
  desirable 
  after 
  every 
  temperature 
  

   observation 
  is 
  also 
  a 
  drawback. 
  Very 
  considerable 
  skill 
  is 
  

   required 
  when 
  the 
  preceding 
  temperature 
  is 
  high, 
  the 
  risk 
  of 
  

   breaking 
  the 
  thermometer 
  being 
  appreciable, 
  and 
  the 
  frequent 
  

   preparation 
  of 
  ice 
  is 
  both 
  troublesome 
  and 
  expensive. 
  The 
  

   conditions 
  under 
  which 
  it 
  is 
  safe 
  to 
  dispense 
  with 
  an 
  ice 
  

   observation 
  and 
  use 
  the 
  Bureau's 
  table 
  of 
  depressed 
  freezing- 
  

   points 
  for 
  verve 
  dar, 
  are 
  somewhat 
  uncertain. 
  The 
  use 
  of 
  the 
  

   table 
  is 
  presumably 
  satisfactory 
  when 
  the 
  temperature 
  has 
  

   been 
  stationary 
  for 
  a 
  long 
  time*, 
  or 
  has 
  been 
  made 
  to 
  rise 
  

   very 
  slowly. 
  But 
  hoiv 
  slow 
  the 
  rise 
  must 
  be, 
  or 
  what 
  is 
  to 
  

   be 
  done 
  when 
  the 
  temperature 
  falls 
  or 
  oscillates, 
  appears 
  by 
  

   no 
  means 
  clear. 
  Again, 
  if 
  the 
  thermometer 
  has 
  been 
  at 
  a 
  

   temperature 
  approaching 
  100° 
  C. 
  one 
  cannot 
  safely 
  plunge 
  it 
  

   straight 
  into 
  ice. 
  The 
  time 
  required 
  adequately 
  to 
  immerse 
  

   it 
  varies 
  with 
  the 
  skill 
  and 
  audacity 
  of 
  the 
  operator. 
  Mean- 
  

   time 
  the 
  thermometer 
  is 
  probably 
  changing 
  its 
  condition, 
  

   with 
  a 
  rapidity 
  which 
  may 
  depend 
  on 
  the 
  time 
  of 
  its 
  exposure 
  

   to 
  the 
  high 
  temperature 
  and 
  its 
  previous 
  history. 
  At 
  really 
  

   high 
  temperatures, 
  150° 
  C. 
  and 
  upwards, 
  increasing 
  difficulties 
  

   can 
  hardly 
  fail 
  to 
  oppose 
  the 
  successful 
  use 
  of 
  the 
  depressed 
  

   zero 
  method, 
  unless 
  the 
  glass 
  employed 
  be 
  extremely 
  little 
  

   affected 
  by 
  sudden 
  physical 
  changes. 
  

  

  In 
  favour 
  of 
  the 
  customary 
  methods 
  of 
  using 
  English 
  glass 
  

   thermometers 
  there 
  are 
  certain 
  advantages. 
  The 
  necessity 
  

   for 
  an 
  internal 
  pressure 
  correction, 
  as 
  we 
  have 
  seen, 
  is 
  avoided 
  

  

  * 
  Schuster 
  and 
  Gannon 
  (Phil. 
  Trans, 
  for 
  1895, 
  p. 
  429) 
  conclude 
  that 
  

   2\ 
  hours 
  exposure 
  of 
  a 
  verve 
  dur 
  thermometer 
  to 
  a 
  temperature 
  of 
  

   22^° 
  C. 
  was 
  insufficient 
  to 
  produce 
  the 
  full 
  depression 
  of 
  zero. 
  The 
  

   subsequent 
  ice-reading 
  was, 
  they 
  think, 
  as 
  much 
  as 
  o, 
  002 
  0. 
  too 
  high. 
  

  

  