﻿526 
  Mr. 
  J. 
  D. 
  Hamilton 
  Dickson 
  on 
  the 
  Reduction 
  to 
  

  

  obtaining 
  the 
  normal 
  air-temperature 
  required. 
  I 
  have 
  given 
  

   reasons 
  * 
  to 
  show 
  that 
  another 
  empiric 
  formula 
  amounting 
  to 
  

   the 
  expansion 
  of 
  the 
  temperature 
  in 
  powers 
  of 
  the 
  electric 
  

   resistance 
  of 
  the 
  wire, 
  while 
  free 
  from 
  some 
  theoretical 
  faults 
  

   attached 
  to 
  other 
  formulae, 
  is 
  at 
  least 
  as 
  good 
  as 
  the 
  compound 
  

   formulae 
  of 
  Prof. 
  Gallendar, 
  and 
  evades 
  the 
  undesirable 
  idea 
  

   of 
  u 
  platinum-temperatures. 
  " 
  

  

  The 
  end 
  I 
  had 
  in 
  view 
  in 
  getting- 
  this 
  formula 
  was 
  the 
  

   reduction 
  to 
  normal 
  air-temperatures 
  of 
  the 
  platinum-tempe- 
  

   ratures 
  in 
  which 
  the 
  results 
  were 
  expressed, 
  of 
  the 
  many 
  and 
  

   important 
  researches 
  of 
  Professors 
  Dewar 
  and 
  Fleming 
  in 
  

   their 
  varied 
  investigations 
  at 
  low 
  temperature. 
  In 
  furtherance 
  

   of 
  this 
  end, 
  Messrs. 
  Petavel 
  and 
  Morris, 
  who 
  so 
  ably 
  assisted 
  

   them 
  in 
  these 
  investigations, 
  told 
  me 
  that 
  the 
  piece 
  of 
  

   platinum-wire 
  which 
  forms 
  the 
  body 
  of 
  the 
  platinum-thermo- 
  

   meter 
  employed 
  by 
  Professors 
  Dewar 
  and 
  Fleming, 
  and 
  

   referred 
  to 
  as 
  " 
  P 
  1? 
  " 
  is 
  the 
  identical 
  piece 
  of 
  wire 
  whose 
  

   variations 
  of 
  electric 
  resistance 
  with 
  temperature 
  are 
  given 
  

   on 
  p. 
  282 
  of 
  Phil. 
  Mag. 
  for 
  Sept. 
  1893, 
  vol. 
  xxxvi.f 
  The 
  

   experiments 
  on 
  this 
  wire 
  there 
  recorded 
  range 
  over 
  nearly 
  

   300°, 
  namely, 
  from 
  about 
  -200° 
  C. 
  to 
  about 
  + 
  100° 
  C. 
  The 
  

   normal 
  air-temperatures 
  observed 
  are 
  given 
  down 
  to 
  the 
  

   freezing-point 
  of 
  water, 
  while 
  lower 
  temperatures 
  are 
  merely 
  

   indicated. 
  Thus, 
  one 
  low 
  temperature 
  which 
  I 
  have 
  employed 
  

   is 
  indicated 
  as 
  " 
  taken 
  in 
  solid 
  carbonic 
  acid 
  and 
  ether/' 
  and 
  

   the 
  other 
  as 
  u 
  taken 
  in 
  liquid 
  oxygen 
  boiling 
  at 
  760 
  millim." 
  

   I 
  have 
  assumed 
  that 
  the 
  former 
  of 
  these 
  temperatures 
  has 
  

   most 
  probably 
  Regnault's 
  value, 
  — 
  78°'2 
  0. 
  ; 
  and 
  there 
  seems 
  

   to 
  be 
  a 
  general 
  convergence 
  of 
  opinion 
  towards 
  — 
  182 
  0, 
  5 
  C. 
  as 
  

   the 
  value 
  of 
  the 
  latter. 
  

  

  With 
  these 
  data, 
  and 
  with 
  the 
  details 
  immediately 
  fol- 
  

   lowing, 
  the 
  annexed 
  table 
  has 
  been 
  constructed, 
  reducing 
  

   the 
  " 
  platinum-temperatures 
  " 
  expressed 
  by 
  means 
  of 
  Pro- 
  

   fessors 
  Dewar 
  and 
  Fleming's 
  platinum-thermometer 
  " 
  P/' 
  to 
  

   normal 
  air-temperatures 
  on 
  the 
  centigrade 
  scale. 
  

  

  * 
  Phil 
  Mag. 
  (1897) 
  xliv. 
  p. 
  445. 
  

  

  t 
  See 
  also 
  this 
  paper, 
  where 
  it 
  is 
  stated 
  that 
  the 
  same 
  wire 
  is 
  used 
  to 
  

   me.xsure 
  the 
  temperatures 
  recorded 
  iu 
  it, 
  p. 
  275. 
  

  

  