﻿390 
  Linebarger 
  — 
  Some 
  Relations 
  between 
  Temperature, 
  

  

  their 
  methods 
  of 
  purification. 
  If 
  it 
  be 
  permitted 
  to 
  judge 
  from 
  

   a 
  single 
  example 
  taken 
  at 
  random, 
  we 
  cannot 
  admit 
  that 
  their 
  

   products 
  were 
  always 
  as 
  pure 
  as 
  necessary 
  ; 
  thus, 
  they 
  found 
  

   for 
  the 
  latent 
  heat 
  of 
  vaporization 
  of 
  ethyl 
  formiate, 
  which, 
  as 
  

   stated 
  just 
  above, 
  Schiff 
  determined 
  to 
  be 
  92*15 
  cal., 
  equal 
  to 
  

   100*4 
  cal. 
  The 
  impurity 
  within 
  compounds 
  of 
  the 
  ether 
  

   class 
  is 
  for 
  the 
  most 
  part 
  water. 
  Since 
  water 
  requires 
  much 
  

   more 
  heat 
  for 
  vaporization 
  than 
  most 
  liquids, 
  its 
  presence, 
  

   even 
  in 
  minimal 
  amount, 
  exercises 
  considerable 
  influence 
  upon 
  

   the 
  value 
  of 
  a 
  determination. 
  In 
  those 
  cases, 
  therefore, 
  where 
  

   water 
  may 
  be 
  present 
  as 
  impurity, 
  the 
  heat 
  of 
  vaporization 
  

   will 
  be 
  too 
  high. 
  And, 
  as 
  a 
  matter 
  of 
  fact, 
  the 
  determinations 
  

   on 
  the 
  ethers 
  made 
  by 
  Andrews, 
  as 
  well 
  as 
  by 
  Berthelot 
  and 
  by 
  

   Ogier, 
  all 
  give 
  values 
  higher 
  than 
  those 
  found 
  by 
  Schiff, 
  who 
  

   took 
  the 
  greatest 
  pains 
  to 
  fully 
  rid 
  his 
  preparations 
  of 
  water. 
  

   The 
  method 
  employed 
  by 
  them 
  is, 
  however, 
  quite 
  beyond 
  any 
  

   but 
  the 
  sharpest 
  criticism, 
  so 
  that 
  their 
  determinations 
  may 
  be 
  

   admitted 
  as 
  sufficiently 
  accurate 
  with 
  the 
  exception 
  of 
  the 
  amyl 
  

   halogen 
  compounds, 
  amylene, 
  ethyl 
  formiate, 
  and 
  sulphur 
  chlor- 
  

   ide. 
  The 
  work 
  of 
  the 
  other 
  investigators 
  may 
  be 
  admitted 
  

   without 
  question, 
  especially 
  that 
  due 
  to 
  Schiff, 
  which 
  is 
  a 
  

   marvel 
  of 
  accuracy. 
  Such 
  determinations 
  as 
  are 
  not 
  trust- 
  

   worthy 
  are 
  marked 
  in 
  the 
  table 
  with 
  a 
  star. 
  

  

  III. 
  

  

  An 
  inspection 
  of 
  Table 
  I 
  shows 
  that 
  the 
  numbers 
  in 
  column 
  

   f 
  are 
  quite 
  constant, 
  with 
  the 
  exception 
  of 
  the 
  alcohols, 
  the 
  

   acids, 
  and 
  the 
  nitro-compounds, 
  as 
  well 
  as 
  water 
  and 
  acetone. 
  

   Leaving 
  these 
  liquids 
  aside 
  for 
  a 
  moment, 
  — 
  their 
  seemingly 
  

   irregular 
  behavior 
  will 
  be 
  explained 
  away 
  later 
  on 
  — 
  we 
  will 
  

   consider 
  the 
  various 
  family 
  of 
  compounds 
  of 
  which 
  Table 
  I 
  is 
  

   made 
  up. 
  Taking 
  all 
  the 
  reliable 
  determinations 
  into 
  con- 
  

   sideration, 
  we 
  find 
  that 
  the 
  average 
  value 
  of 
  the 
  u 
  constant" 
  is 
  

   for 
  about 
  seventy 
  liquids 
  equal 
  to 
  £0 
  m 
  70, 
  the 
  greatest 
  value 
  

   being 
  22-04 
  for 
  bromine 
  (Andrews 
  I).* 
  For 
  the 
  elements 
  and 
  

   inorganic 
  compounds, 
  the 
  " 
  constant 
  " 
  is 
  equal 
  to 
  W-J^l 
  with 
  

   22*04 
  and 
  19*66 
  as 
  extreme 
  values 
  ; 
  for 
  the 
  hydrocarbons, 
  to 
  

   W' 
  19, 
  20-63 
  and 
  19*58 
  being 
  the 
  extreme 
  valnes 
  ; 
  for 
  the 
  halogen 
  

   compounds, 
  to 
  W'63, 
  with 
  extreme 
  values 
  equal 
  to 
  21*16 
  and 
  

   19-59; 
  for 
  the 
  esters, 
  to 
  20-87, 
  the 
  extremes 
  being 
  21*43 
  and 
  

   20*36. 
  With 
  the 
  exception 
  of 
  the 
  esters, 
  the 
  determinations 
  

   have 
  been 
  made 
  by 
  different 
  men 
  in 
  different 
  ways, 
  so 
  that 
  a 
  

   great 
  degree 
  of 
  " 
  constancy 
  " 
  is 
  hardly 
  to 
  be 
  expected 
  ; 
  yet 
  the 
  

  

  * 
  The 
  determination 
  by 
  Berthelot 
  and 
  Ogier 
  (xiv), 
  however, 
  gives 
  20-95 
  as 
  the 
  

   value 
  of 
  the 
  " 
  constant," 
  so 
  that 
  it 
  is 
  perhaps 
  better 
  to 
  reject 
  Andrews' 
  determi- 
  

   nation. 
  If 
  that 
  be 
  done 
  the 
  greatest 
  value 
  is 
  2T54 
  for 
  methylal 
  (Berthelot) 
  and 
  

   the 
  smallest 
  value 
  being 
  19-58 
  for 
  pseudocumene 
  (Schiff). 
  

  

  