﻿W. 
  P. 
  White 
  — 
  Silicate 
  Specific 
  Heats. 
  31 
  

  

  prompt 
  a-p 
  quartz 
  and 
  cristobalite 
  inversions. 
  In 
  other 
  

   cases 
  the 
  rearrangement 
  of 
  atoms 
  seems 
  to 
  be 
  too 
  com- 
  

   plex 
  to 
  take 
  place 
  all 
  at 
  once, 
  and 
  there 
  is 
  no 
  general 
  

   shifting 
  at 
  all, 
  but 
  the 
  new 
  crystals 
  grow 
  at 
  the 
  expense 
  

   of 
  the 
  old 
  as 
  they 
  might 
  grow 
  in 
  glass, 
  only 
  often 
  more 
  

   slowly. 
  The 
  final 
  result 
  is 
  then 
  usually 
  an 
  aggregate 
  of 
  

   small 
  crystals. 
  The 
  wollastonite-to-pseudo-wollastonite 
  

   and 
  quartz-to-cristobalite 
  inversions 
  proceed 
  in 
  this 
  way. 
  

   It 
  does 
  not 
  seem 
  necessary 
  to 
  suppose 
  that 
  these 
  two 
  

   kinds 
  of 
  inversion 
  are 
  fundamentally 
  different, 
  to 
  sup- 
  

   pose, 
  for 
  instance, 
  that 
  one 
  is 
  "chemical," 
  the 
  other 
  

   "physical," 
  or 
  that 
  each 
  comes 
  from 
  a 
  change 
  in 
  a 
  dif- 
  

   ferent 
  kind 
  of 
  linkage. 
  This 
  may 
  be 
  the 
  case, 
  of 
  course, 
  

   but 
  the 
  mere 
  amount 
  of 
  change 
  may 
  determine 
  the 
  way 
  

   the 
  change 
  proceeds. 
  

  

  Fenner, 
  in 
  his 
  comprehensive 
  study 
  of 
  the 
  various 
  silica 
  

   inversions, 
  39 
  puts 
  the 
  a-p 
  inversions 
  in 
  a 
  different 
  class 
  

   from 
  the 
  quartz- 
  tridymite, 
  etc., 
  inversions 
  on 
  the 
  ground 
  

   of 
  their 
  greater 
  promptness 
  and 
  the 
  smaller 
  change 
  in 
  

   crystallographic 
  properties 
  involved. 
  A 
  still 
  more 
  fun- 
  

   damental 
  reason 
  for 
  distinguishing 
  seems 
  to 
  be 
  the 
  fact 
  

   that 
  the 
  two 
  kinds 
  of 
  inversion 
  are, 
  so 
  to 
  speak, 
  super- 
  

   posed 
  upon 
  one 
  another; 
  that 
  is, 
  the 
  prompt 
  inversion 
  

   remains 
  when 
  quartz 
  changes 
  to 
  tridymite 
  or 
  cristobal- 
  

   ite, 
  40 
  and 
  conversely, 
  its 
  occurrence 
  does 
  not 
  in 
  the 
  least 
  

   change 
  either 
  of 
  these 
  more 
  permanent 
  forms. 
  Different 
  

   mechanisms, 
  therefore, 
  determine 
  the 
  different 
  sorts 
  of 
  

   change, 
  though 
  these 
  mechanisms 
  can 
  affect 
  each 
  other, 
  

   since 
  the 
  prompt 
  inversion 
  has 
  a 
  somewhat 
  different 
  

   character 
  in 
  each 
  of 
  the 
  three 
  permanent 
  forms. 
  

  

  Our 
  knowledge 
  of 
  crystal 
  atomic 
  patterns 
  does 
  not 
  

   seem 
  to 
  be 
  extensive 
  enough 
  at 
  present 
  to 
  permit 
  develop- 
  

   ing 
  a 
  theory 
  of 
  inversion 
  phenomena 
  out 
  of 
  considera- 
  

   tions 
  like 
  those 
  first 
  given. 
  A 
  theory 
  has 
  been 
  formulated 
  

   which 
  attempts 
  to 
  correlate 
  inversions 
  with 
  the 
  equi- 
  

   librium 
  changes 
  which 
  have 
  been 
  studied 
  in 
  liquids, 
  or 
  in 
  

   liquid-and-solid 
  systems. 
  41 
  This 
  theory 
  (or 
  hypothesis) 
  

  

  39 
  Loc. 
  cit. 
  

  

  40 
  Probably, 
  that 
  is, 
  it 
  is 
  not 
  yet 
  proved 
  that 
  these 
  prompt 
  inversions 
  are 
  

   not 
  of 
  different 
  nature, 
  whose 
  occurrence 
  where 
  they 
  are 
  is 
  a 
  coincidence, 
  

   but 
  this 
  is 
  unlikely. 
  

  

  41 
  A. 
  Smits, 
  Nouvelle 
  ThSorie 
  du 
  Phenomene 
  de 
  1 
  'Allotropie, 
  Kevue 
  

   gerierale 
  des 
  Sciences, 
  15 
  March, 
  1914; 
  Zs. 
  Phys. 
  Chem., 
  76, 
  421, 
  1911. 
  

  

  G. 
  Tammann, 
  Zur 
  Atomistischen 
  Theorie 
  des 
  Polymorphismus, 
  Zs. 
  Phys. 
  

   Chem., 
  82, 
  179, 
  1913. 
  

  

  