﻿Fenner 
  — 
  Stability 
  Relations 
  of 
  Silica 
  Minerals. 
  367 
  

  

  has 
  been 
  formed 
  at 
  a 
  very 
  high 
  temperature 
  it 
  will 
  give 
  a 
  peak 
  

   on 
  the 
  heating 
  curve 
  (such 
  as 
  shown 
  in 
  fig. 
  6) 
  at 
  about 
  270°, 
  

   and 
  on 
  the 
  cooling 
  curve 
  at 
  about 
  240° 
  ; 
  if 
  formed 
  at 
  succes- 
  

   sively 
  lower 
  temperatures 
  these 
  two 
  points 
  will 
  drop 
  step 
  by 
  

   step 
  until 
  at 
  the 
  lowest 
  limit 
  at 
  which 
  cristobalite 
  has 
  been 
  pro- 
  

   duced 
  the 
  corresponding 
  points 
  are 
  220° 
  and 
  198°, 
  a 
  drop 
  of 
  

   about 
  50° 
  and 
  40° 
  respectively. 
  

  

  If 
  cristobalite 
  is 
  of 
  the 
  pseudo-unary 
  but 
  actually 
  binary 
  

   nature 
  which 
  has 
  been 
  suggested, 
  the 
  temperature-composition 
  

   diagram 
  should 
  be 
  of 
  the 
  general 
  form 
  shown 
  in 
  fig. 
  7. 
  

  

  In 
  this 
  the 
  two 
  kinds 
  of 
  molecules 
  are 
  represented 
  by 
  

   \ 
  and 
  p. 
  The 
  proportions 
  of 
  these 
  in 
  equilibrium 
  at 
  different 
  

   temperatures 
  from 
  A 
  to 
  B 
  are 
  given 
  by 
  the 
  line 
  AB, 
  and 
  the 
  

   /3-a 
  inversions 
  of 
  the 
  pseudo-unary 
  mix-crystals 
  are 
  given 
  by 
  

   the 
  curves 
  CBD 
  and 
  CED. 
  In 
  the 
  inversion 
  of 
  a 
  /3-crystal 
  of 
  

   the 
  composition 
  B 
  a 
  small 
  amount 
  of 
  a-crystal 
  of 
  the 
  composi- 
  

   tion 
  E 
  is 
  first 
  formed 
  and 
  by 
  the 
  continuation 
  of 
  inversion 
  the 
  

   composition 
  of 
  the 
  /3-crystal 
  moves 
  along 
  BG 
  and 
  of 
  the 
  

   a-crystal 
  along 
  EF. 
  At 
  F 
  the 
  whole 
  has 
  been 
  inverted. 
  This 
  

   diagram, 
  however, 
  represents 
  merely 
  a 
  theoretical 
  conception 
  

   involving 
  the 
  requirement 
  that 
  equilibrium 
  keeps 
  pace 
  with 
  

   changes 
  of 
  temperature. 
  "With 
  such 
  a 
  substance 
  as 
  cristobalite 
  

   this 
  requirement 
  cannot, 
  of 
  course, 
  be 
  strictly 
  fulfilled 
  and 
  the 
  

   process 
  will 
  vary 
  in 
  accordance. 
  The 
  composition 
  of 
  the 
  

   material 
  formed 
  at 
  any 
  temperature 
  such 
  as 
  A 
  or 
  H 
  will 
  be 
  

   reached 
  only 
  after 
  very 
  long 
  heating, 
  if 
  at 
  all. 
  It 
  is 
  found, 
  in 
  

   fact, 
  that 
  the 
  actual 
  composition 
  (that 
  is, 
  relative 
  proportions 
  

   of 
  X 
  and 
  /x 
  molecules) 
  of 
  the 
  material 
  formed 
  at 
  any 
  given 
  

   temperature 
  depends 
  upon 
  the 
  nature 
  of 
  the 
  material 
  with 
  

   which 
  we 
  start. 
  If 
  quartz 
  is 
  employed 
  at 
  the 
  temperature 
  H 
  

   a 
  composition 
  I 
  is 
  attained, 
  if 
  amorphous 
  silica, 
  a 
  composition 
  K. 
  

   This 
  is 
  not 
  surprising, 
  if 
  we 
  consider 
  that 
  in 
  the 
  one 
  case 
  we 
  

   are 
  approaching 
  the 
  composition 
  H 
  from 
  a 
  lower 
  form, 
  and 
  in 
  

   the 
  second 
  case 
  from 
  a 
  higher. 
  The 
  inference 
  is 
  that 
  the 
  

   quartz 
  molecule, 
  in 
  breaking 
  up, 
  forms 
  at 
  first 
  relatively 
  more 
  

   of 
  the 
  lower 
  (X) 
  molecule 
  than 
  does 
  amorphous 
  silica 
  at 
  the 
  

   same 
  temperature. 
  The 
  velocity 
  of 
  reaction 
  in 
  converting 
  one 
  

   kind 
  of 
  molecule 
  into 
  another 
  is 
  so 
  slow 
  (as 
  are 
  all 
  changes 
  

   throughout 
  the 
  silica 
  diagram 
  which 
  involve 
  the 
  formation 
  of 
  

   new 
  molecules) 
  that 
  the 
  composition 
  which, 
  at 
  the 
  temperature 
  

   H, 
  should 
  be 
  represented 
  by 
  the 
  point 
  H 
  on 
  the 
  AB 
  line, 
  is 
  

   actually 
  represented 
  by 
  the 
  point 
  I 
  in 
  one 
  case 
  and 
  the 
  point 
  

   K 
  in 
  the 
  other. 
  Moreover, 
  after 
  the 
  material 
  has 
  once 
  been 
  

   formed 
  and 
  its 
  composition 
  fixed 
  it 
  retains 
  it 
  during 
  cooling. 
  

   Indeed 
  it 
  is 
  from 
  this 
  fact 
  that 
  differences 
  of 
  composition 
  can 
  

   be 
  recognized, 
  for 
  material 
  of 
  the 
  composition 
  K', 
  when 
  

   cooled, 
  will 
  still 
  retain 
  this 
  composition 
  (instead 
  of 
  changing 
  

   along 
  K'Bj 
  and 
  will 
  give 
  the 
  ft-a 
  inversion 
  corresponding 
  to 
  

  

  