﻿14:2 
  C. 
  Barns 
  — 
  Continuity 
  of 
  Solid 
  and 
  Liquid. 
  

  

  which 
  lies 
  at 
  the 
  root 
  of 
  all 
  manifestations 
  of 
  hysteresis, 
  

   whether 
  observed 
  electrically 
  (Cohn, 
  Ewing, 
  Schumann), 
  or 
  

   magnetically 
  (Warburg, 
  Ewing), 
  or 
  as 
  a 
  purely 
  mechanical 
  

   result 
  in 
  my 
  work,* 
  during 
  fusion, 
  as 
  above, 
  during 
  solution, 
  

   § 
  29, 
  etc. 
  

  

  24. 
  James 
  Thomson's 
  double 
  inflections. 
  — 
  Solidification 
  al- 
  

   most 
  always 
  sets 
  in 
  at 
  once. 
  One 
  would 
  expect 
  this 
  : 
  for 
  if 
  

   there 
  be 
  condensation 
  or 
  crystallization 
  at 
  any 
  one 
  point, 
  it 
  will 
  

   form 
  the 
  nucleus 
  from 
  which 
  the 
  whole 
  column 
  will 
  be 
  solidi- 
  

   fied, 
  so 
  far 
  as 
  it 
  lies 
  in 
  the 
  field 
  of 
  volume 
  lag. 
  Only 
  in 
  one 
  

   case 
  (liquid 
  isothermal 
  83° 
  60 
  m 
  to 
  63 
  m 
  ) 
  did 
  I 
  obtain 
  evidence 
  of 
  

   curvature. 
  Usually 
  even 
  at 
  low 
  temperatures 
  the 
  path 
  is 
  pre- 
  

   cipitous, 
  because 
  pressure 
  cannot 
  be 
  lowered 
  rapidly 
  enough. 
  

  

  The 
  reverse 
  of 
  this 
  holds 
  in 
  case 
  of 
  fusion. 
  Here 
  the 
  

   initial 
  or 
  stable 
  contours 
  of 
  James 
  Thomson's 
  circumflexures 
  

   are 
  well 
  marked. 
  It 
  is 
  true 
  that 
  fusion 
  cannot 
  take 
  place 
  in- 
  

   stantaneously, 
  because 
  heat 
  cannot 
  be 
  supplied 
  fast 
  enough. 
  

   It 
  is 
  also 
  true 
  that 
  if 
  temperature 
  be 
  not 
  quite 
  identical 
  

   throughout 
  the 
  length 
  of 
  column, 
  fusion 
  will 
  first 
  take 
  place 
  at 
  

   the 
  hotter 
  planes 
  below, 
  and 
  proceed 
  thence 
  to 
  the 
  top.f 
  In 
  the 
  

   present 
  experiments, 
  however, 
  the 
  phenomenon 
  occurs 
  with 
  

   the 
  same 
  uniformity 
  at 
  all 
  temperatures, 
  and 
  is 
  quite 
  pro- 
  

   nounced 
  in 
  the 
  steam 
  bath. 
  § 
  4. 
  Hence, 
  taking 
  into 
  additional 
  

   consideration 
  the 
  evidence 
  of 
  § 
  22, 
  1 
  conclude 
  that 
  the 
  initial 
  

   contours 
  are 
  static 
  and 
  regard 
  them 
  as 
  partially 
  evidencing 
  

   James 
  Thomson's:}; 
  well 
  known 
  inference 
  relative 
  to 
  the 
  doubly 
  

   inflected 
  contours 
  of 
  the 
  isothermal 
  paths 
  accompanying 
  

   change 
  of 
  physical 
  state. 
  When 
  fusion 
  actually 
  sets 
  in, 
  the 
  

   phenomenon 
  is 
  no 
  longer 
  observable 
  ; 
  for 
  the 
  physical 
  parts 
  of 
  

   the 
  substance 
  now 
  exist 
  in 
  widely 
  different 
  thermal 
  states. 
  In 
  

   figure 
  2 
  the 
  full 
  contours 
  are 
  indicated 
  by 
  dotted 
  lines. 
  

  

  25. 
  The 
  characteristic 
  specific 
  volumes. 
  — 
  Mere 
  inspection 
  of 
  

   the 
  chart, 
  figure 
  2, 
  shows 
  that 
  the 
  volume 
  at 
  which 
  solidifica- 
  

   tion 
  takes 
  place, 
  decreases 
  as 
  temperature 
  increases, 
  while 
  the 
  

   volume 
  into 
  which 
  the 
  substance 
  solidifies 
  either 
  increases 
  or 
  

   remains 
  stationary 
  in 
  value. 
  In 
  table 
  10, 
  I 
  have 
  inscribed 
  the 
  

   corresponding 
  values 
  of 
  pressure 
  and 
  of 
  volume, 
  observed 
  at 
  

   the 
  solidification 
  points, 
  in 
  each 
  of 
  my 
  four 
  sets 
  of 
  results.§ 
  

   The 
  data 
  are 
  plotted 
  in 
  figure 
  3, 
  the 
  volumes 
  being 
  abscissas, 
  

   the 
  pressures 
  ordinates. 
  To 
  distinguish 
  the 
  points 
  of 
  this 
  

   diagram, 
  they 
  are 
  surrounded 
  by 
  little 
  circles, 
  to 
  which 
  the 
  

   number 
  of 
  the 
  series 
  is 
  attached. 
  

  

  A 
  similar 
  and 
  equally 
  expressive 
  table 
  may 
  be 
  deduced 
  by 
  

   finding 
  the 
  characteristic 
  volumes 
  at 
  the 
  successive 
  melting 
  

  

  * 
  Cf. 
  my 
  results 
  on 
  the 
  Bourdon 
  gauge 
  in 
  a 
  current 
  number 
  of 
  the 
  Phil. 
  Mag. 
  

   f 
  I 
  have 
  actually 
  observed 
  this 
  in 
  glass 
  capillary 
  tubes, 
  when 
  the 
  vapor 
  baths 
  

   were 
  imperfect. 
  \ 
  James 
  Thomson 
  : 
  Phil. 
  Mag. 
  (TV), 
  xlii, 
  p. 
  227, 
  1872. 
  

  

  § 
  The 
  above 
  tables 
  and 
  figure 
  2 
  exhibiting 
  but 
  one 
  of 
  these 
  sets. 
  

  

  