﻿372 
  Mr. 
  E. 
  A. 
  Lehfeldt 
  on 
  a 
  Numerical 
  Evaluation 
  of 
  

  

  obvious 
  that 
  no 
  reliance 
  can 
  be 
  put 
  in 
  deductions 
  made 
  from 
  

   the 
  mixtures 
  containing 
  much 
  air, 
  and 
  that 
  the 
  whole 
  question 
  

   is 
  in 
  a 
  very 
  unsatisfactory 
  state. 
  The 
  high 
  temperature 
  

   measurements 
  gave 
  6 
  = 
  0° 
  155, 
  and 
  the 
  mean 
  between 
  that 
  

   and 
  0*108 
  reduced 
  to 
  the 
  atmosphere 
  as 
  unit 
  of 
  pressure 
  gives 
  

   e= 
  +0 
  o, 
  039 
  : 
  which 
  is 
  the 
  number 
  used 
  by 
  Lord 
  Kelvin 
  in 
  

   the 
  Encyclopaedia 
  Brittanica 
  article. 
  It 
  should 
  be 
  noted, 
  

   however, 
  that 
  this 
  appears 
  to 
  indicate 
  a 
  greater 
  effect 
  at 
  the 
  

   higher 
  temperature, 
  contrary 
  to 
  what 
  was 
  observed 
  for 
  the 
  

   other 
  gases. 
  

  

  Air. 
  — 
  In 
  their 
  earlier 
  memoir, 
  Joule 
  and 
  Thomson* 
  

   record 
  experiments 
  with 
  air 
  at 
  17° 
  and 
  at 
  91°, 
  yielding 
  

   — 
  o, 
  259 
  and 
  — 
  0°*206 
  per 
  atmosphere 
  respectively. 
  In 
  the 
  

   later 
  memoir 
  f 
  they 
  paid 
  more 
  attention 
  to 
  the 
  experiments 
  

   at 
  high 
  temperatures, 
  so 
  probably 
  the 
  results 
  are 
  more 
  trust- 
  

   worthy 
  than 
  that 
  first 
  found 
  for 
  91°. 
  The 
  following 
  table 
  shows 
  

   the 
  mean 
  of 
  the 
  observations 
  grouped 
  about 
  four 
  different 
  

   temperatures, 
  and 
  also 
  the 
  value 
  of 
  e 
  calculated 
  on 
  the 
  

   assumption 
  that 
  it 
  varies 
  inversely 
  as 
  the 
  square 
  of 
  the 
  

   absolute 
  temperature 
  : 
  — 
  

  

  Mean 
  

   temperature. 
  

  

  No. 
  of 
  

   experiments. 
  

  

  e 
  (obs.). 
  

  

  e 
  (calc). 
  

  

  o 
  

   7 
  

  

  26 
  

  

  50 
  

  

  93 
  

  

  8 
  

   2 
  

   6 
  

  

  -0 
  ? 
  263 
  

   0-229 
  

   0-209 
  

   0-152 
  

  

  -0°261 
  > 
  

   0-229 
  

   0196 
  

   0153, 
  

  

  > 
  per 
  

   atmo. 
  

  

  The 
  mean 
  over 
  the 
  range 
  0° 
  to 
  100° 
  is 
  — 
  0°200. 
  Lord 
  

   Kelvin, 
  in 
  the 
  Encycl. 
  Britt., 
  gives 
  the 
  value 
  — 
  0°\208, 
  which 
  

   is 
  practically 
  the 
  same. 
  

  

  Nitrogen. 
  — 
  The 
  single 
  experiment 
  made 
  at 
  7° 
  with 
  this 
  gas 
  

   gave 
  a 
  cooling 
  of 
  l 
  o, 
  034 
  for 
  100 
  inches 
  of 
  mercury 
  ; 
  and 
  two 
  

   experiments 
  at 
  92° 
  gave 
  0*576 
  and 
  0*691 
  respectively. 
  

   These 
  numbers 
  are 
  fairly 
  in 
  agreement 
  with 
  the 
  assumption 
  

   that 
  the 
  effect 
  varies 
  inversely 
  as 
  the 
  square 
  of 
  the 
  absolute 
  

   temperature, 
  and 
  lead 
  to 
  

  

  e 
  = 
  - 
  0°-332 
  per 
  atmo 
  at 
  0°, 
  

   e 
  = 
  - 
  0°'224 
  „ 
  mean 
  between 
  0° 
  and 
  100°. 
  

  

  For 
  carbon 
  dioxide, 
  Joule 
  and 
  Thomson 
  state 
  the 
  same 
  

  

  * 
  Phil. 
  Trans. 
  1854. 
  f 
  Ibid. 
  1862. 
  

  

  