﻿Dr. 
  C. 
  Chree's 
  Notes 
  on 
  Thermometry. 
  225 
  

  

  at 
  —39° 
  C, 
  and 
  Balfour 
  Stewart's 
  * 
  determination 
  of 
  the 
  

   freezing-point 
  of 
  mercury 
  on 
  the 
  scale 
  of 
  the 
  air-thermometer, 
  

   viz. 
  -37°-93 
  F. 
  (or 
  -38°-85 
  C), 
  differs 
  by 
  only 
  0°'05 
  C. 
  

   from 
  the 
  Bureau's 
  determination 
  on 
  the 
  hydrogen 
  scale. 
  

   This 
  collateral 
  evidence 
  is, 
  however, 
  somewhat 
  weakened 
  by 
  

   the 
  fact 
  that 
  Balfour 
  Stewart 
  states 
  that 
  the 
  mercury 
  used 
  in 
  

   the 
  experiments 
  on 
  thermometer 
  No. 
  45 
  was 
  not 
  the 
  same 
  as 
  

   that 
  used 
  with 
  the 
  air-thermometer. 
  

  

  § 
  18. 
  Since 
  Balfour 
  Stewart's 
  investigation 
  there 
  have 
  

   been 
  numerous 
  observations 
  of 
  the 
  freezing-point 
  of 
  mercury 
  

   on 
  calibrated 
  Kew 
  standard 
  thermometers 
  of 
  English 
  glass. 
  

   The 
  results 
  deduced 
  at 
  Kew 
  Observatory 
  for 
  the 
  freezing- 
  

   point 
  of 
  mercury 
  on 
  the 
  natural 
  (fixed 
  zero) 
  scales 
  of 
  twenty 
  

   of 
  these 
  thermometers 
  are 
  as 
  follows 
  : 
  — 
  

  

  highest 
  -38°'15F. 
  (-38°-97C), 
  

   mean 
  -38°-35 
  F. 
  (-39°'08 
  C), 
  

   lowest 
  -38°-50F. 
  (-39°-17C). 
  

  

  The 
  thermometers 
  were 
  read 
  to 
  the 
  nearest 
  o, 
  05 
  F. 
  only. 
  

  

  In 
  sixteen 
  out 
  of 
  the 
  twenty 
  thermometers 
  the 
  reading 
  lay 
  

   between 
  — 
  38°"25 
  F. 
  and 
  — 
  38 
  0, 
  45 
  F., 
  and 
  as 
  errors 
  of 
  reading 
  

   of 
  o, 
  05 
  F. 
  might 
  easily 
  arise, 
  there 
  can 
  be 
  little 
  doubt 
  that 
  the 
  

   mean 
  found 
  above 
  is 
  fairly 
  representative 
  for 
  ordinary 
  English 
  

   glass. 
  

  

  To 
  test 
  the 
  accuracy 
  of 
  the 
  method, 
  and 
  obtain 
  a 
  connexion 
  

   with 
  the 
  hydrogen 
  scale, 
  I 
  had 
  three 
  of 
  the 
  twenty 
  thermo- 
  

   meters 
  — 
  which 
  had 
  remained 
  at 
  the 
  Observatory 
  — 
  partially 
  

   recalibrated 
  and 
  retried 
  together 
  in 
  the 
  mercury 
  bath, 
  while 
  

   two 
  independent 
  observations 
  were 
  made 
  in 
  the 
  bath 
  with 
  a 
  

   Tonnelot 
  thermometer 
  standardized 
  at 
  the 
  Bureau 
  Inter- 
  

   national. 
  The 
  results 
  obtained 
  with 
  the 
  three 
  Kew 
  standards, 
  

   after 
  allowing 
  for 
  the 
  secular 
  change 
  of 
  zero, 
  were 
  practically 
  

   identical 
  with 
  those 
  found 
  on 
  separate 
  occasions 
  many 
  years 
  

   ago. 
  While 
  the 
  results 
  obtained 
  on 
  the 
  two 
  occasions 
  with 
  

   the 
  Tonnelot 
  thermometer, 
  read 
  to 
  o, 
  01 
  C. 
  by 
  the 
  unaided 
  

   eye, 
  differed 
  by 
  only 
  o, 
  02 
  C, 
  and 
  gave 
  for 
  the 
  freezing-point 
  

   of 
  mercury 
  

  

  — 
  39 
  0, 
  286 
  C. 
  on 
  the 
  verve 
  dur 
  scale, 
  

   or, 
  applying 
  the 
  Bureau's 
  table, 
  

  

  — 
  38°*86 
  C. 
  on 
  the 
  hydrogen 
  scale. 
  

  

  The 
  reading 
  with 
  the 
  Tonnelot 
  thermometer 
  is 
  o, 
  06 
  C. 
  

   lower 
  than 
  that 
  accepted 
  by 
  the 
  Bureau, 
  and 
  1 
  am 
  not 
  certain 
  

  

  * 
  L. 
  c. 
  p. 
  435. 
  

  

  