﻿316 
  Dr. 
  C. 
  Chree's 
  Notes 
  on 
  Thermometry. 
  

  

  of 
  the 
  same 
  type 
  as 
  (27), 
  but 
  have 
  taken 
  for 
  t' 
  the 
  temper- 
  

   ature 
  of 
  the 
  room, 
  and 
  determined 
  C 
  for 
  each 
  thermometer 
  

   by 
  special 
  experiment. 
  These 
  experiments 
  have 
  been 
  made 
  

   in 
  two 
  ways 
  : 
  — 
  

  

  1. 
  The 
  lowering 
  in 
  the 
  boiling-point 
  reading 
  caused 
  by 
  

   increasing 
  the 
  length 
  of 
  mercury-column 
  emergent 
  from 
  

   the 
  hypsometer 
  has 
  been 
  observed 
  for 
  several 
  lengths 
  of 
  

   column 
  *. 
  

  

  2, 
  The 
  thermometer 
  has 
  been 
  attached 
  to 
  a 
  platinum 
  

   resistance 
  pyrometer, 
  so 
  that 
  the 
  centres 
  of 
  the 
  bulb 
  and 
  

   resistance- 
  coil 
  are 
  at 
  the 
  same 
  level 
  in 
  a 
  well-stirred 
  bath 
  of 
  

   fusible 
  metal. 
  Then 
  two 
  series 
  of 
  simultaneous 
  readings 
  

   have 
  been 
  taken 
  of 
  the 
  two 
  instruments, 
  the 
  immersions 
  in 
  

   the 
  two 
  series 
  being 
  different. 
  If 
  we 
  assume 
  that 
  the 
  

   difference 
  of 
  emergence 
  has 
  not 
  affected 
  the 
  readings 
  of 
  the 
  

   platinum 
  pyrometer, 
  whose 
  coil 
  is 
  completely 
  immersed 
  in 
  

   both 
  series 
  of 
  experiments, 
  one 
  has 
  obviously 
  the 
  means 
  of 
  

   calculating 
  C. 
  

  

  The 
  first 
  of 
  the 
  above 
  methods 
  enables 
  one 
  to 
  check 
  the 
  

   adequacy 
  of 
  the 
  formula 
  for 
  a 
  series 
  of 
  values 
  of 
  t 
  1 
  at 
  a 
  

   specified 
  value 
  of 
  t 
  ; 
  the 
  latter 
  enables 
  one 
  to 
  test 
  it 
  readily 
  

   for 
  a 
  series 
  of 
  values 
  of 
  t. 
  The 
  second 
  method 
  is 
  by 
  no 
  

   means 
  very 
  difficult 
  in 
  practice 
  when 
  one 
  is 
  content 
  with 
  

   accuracy 
  of 
  the 
  order 
  o, 
  l 
  C. 
  in 
  the 
  readings, 
  and 
  some 
  

   promising 
  results 
  have 
  been 
  obtained 
  by 
  means 
  of 
  it. 
  There 
  

   are 
  several 
  possible 
  pitfalls, 
  however, 
  which 
  want 
  fuller 
  

   investigation. 
  

  

  When 
  the 
  resistance-coil 
  of 
  a 
  platinum 
  pyrometer 
  is 
  at 
  

   a 
  high 
  temperature, 
  there 
  can 
  hardly 
  fail 
  to 
  be 
  an 
  appreciable 
  

   transfer 
  of 
  heat 
  up 
  the 
  tube, 
  both 
  by 
  air 
  convection 
  currents 
  

   and 
  conduction 
  along 
  the 
  metal 
  leads, 
  tending 
  of 
  course 
  to 
  

   depress 
  the 
  reading 
  of 
  the 
  pyrometer, 
  and 
  conceivably 
  this 
  

   depression 
  may 
  vary 
  sensibly 
  when 
  the 
  length 
  of 
  stem 
  

   immersed 
  above 
  the 
  platinum 
  spiral 
  is 
  altered. 
  Again 
  at 
  

   high 
  temperatures 
  there 
  is 
  sometimes 
  difficulty 
  in 
  securing 
  a 
  

   sufficiently 
  slow 
  steady 
  rise 
  of 
  temperature, 
  so 
  that 
  sensible 
  

   error 
  may 
  arise 
  from 
  a 
  difference 
  in 
  the 
  lag 
  of 
  instruments 
  

   so 
  different 
  in 
  type 
  as 
  mercury 
  and 
  platinum 
  resistance 
  

   thermometers. 
  

  

  The 
  apparatus 
  employed 
  in 
  the 
  second 
  method 
  was 
  kindly 
  

   lent 
  to 
  the 
  Kew 
  Observatory 
  by 
  Mr. 
  C. 
  T. 
  Hey 
  cock 
  and 
  

   Mr. 
  F. 
  H. 
  Neville, 
  who 
  likewise 
  gave 
  me 
  the 
  benefit 
  of 
  their 
  

   varied 
  experience 
  in 
  comparing 
  thermometers 
  by 
  means 
  

   of 
  it. 
  

  

  * 
  

  

  Cf. 
  Mills, 
  Trans. 
  K, 
  b\ 
  E. 
  vol. 
  xxix. 
  1880, 
  p. 
  569, 
  

  

  