﻿Washington 
  — 
  Analyses 
  of 
  Italian 
  Volcanic 
  Rocks. 
  45 
  

  

  I. 
  Ciminite. 
  La 
  Colonetta, 
  Monte 
  Ciraino. 
  

   II. 
  Ciminite. 
  Fontana 
  Fiescoli 
  * 
  Monte 
  Cimino. 
  Jour. 
  Geol., 
  

  

  iv, 
  p. 
  837, 
  1896. 
  

   III. 
  Ciminite. 
  Fontana 
  Fiescoli, 
  Monte 
  Cimino. 
  New 
  analysis. 
  

   The 
  analysis 
  (I) 
  shows 
  very 
  clearly 
  the 
  intermediate 
  char- 
  

   acter 
  of 
  these 
  rocks 
  in 
  its 
  medium 
  Si0 
  2 
  , 
  rather 
  low 
  A1 
  2 
  3 
  , 
  and 
  

   high 
  MgO, 
  CaO 
  and 
  K 
  2 
  0, 
  which, 
  taken 
  together, 
  are 
  neither 
  

   characteristic 
  of 
  an 
  andesite 
  or 
  basalt 
  on 
  the 
  one 
  hand 
  nor 
  of 
  a 
  

   trachyte 
  on 
  the 
  other. 
  Since 
  the 
  rock 
  is 
  not 
  noncrystalline 
  a 
  

   determination 
  of 
  the 
  relative 
  mineralogical 
  composition 
  is 
  not 
  

   perfectly 
  sure, 
  but 
  a 
  calculation 
  based 
  on 
  the 
  analysis 
  indicates 
  

   that, 
  if 
  the 
  crystallization 
  had 
  been 
  complete, 
  the 
  composition 
  

   would 
  have 
  been 
  as 
  follows 
  : 
  

  

  Orthoclase 
  _ 
  37'9 
  

  

  Labradorite 
  (Ab 
  4 
  An 
  5 
  ) 
  .. 
  26*5 
  

  

  Diopside 
  _ 
  16*5 
  

  

  Olivine 
  17*3 
  

  

  Magnetite.. 
  _ 
  _.. 
  18 
  

  

  100-0 
  

  

  For 
  comparison 
  .the 
  original 
  type 
  analysis 
  is 
  reproduced 
  in 
  

   II. 
  It 
  will 
  be 
  seen 
  that 
  the 
  two 
  resemble 
  each 
  other 
  in 
  all 
  

   particulars, 
  except 
  in 
  A1 
  2 
  3 
  and 
  MgO, 
  which 
  are 
  respectively 
  

   lower 
  and 
  higher 
  in 
  the 
  latest 
  analysis. 
  Suspicion 
  was 
  natu- 
  

   rally 
  aroused 
  that 
  in 
  the 
  older 
  analysis, 
  which, 
  it 
  may 
  be 
  men- 
  

   tioned, 
  was 
  the 
  first 
  complete 
  rock 
  analysis 
  made 
  by 
  me, 
  part 
  

   of 
  the 
  MgO 
  had 
  been 
  thrown 
  down 
  with 
  the 
  A1 
  2 
  3 
  and 
  Fe 
  2 
  3 
  . 
  

   This 
  is 
  a 
  mistake 
  which 
  is 
  frequently 
  made 
  by 
  chemists 
  inex- 
  

   perienced 
  in 
  rock 
  analysis, 
  and 
  one 
  to 
  which 
  attention 
  has 
  been 
  

   called 
  by 
  Pirsson,f 
  HillebrandJ 
  and 
  myself. 
  § 
  

  

  This 
  supposition 
  has 
  been 
  confirmed 
  by 
  a 
  partial 
  analysis, 
  

   recently 
  made, 
  with 
  an 
  especially 
  careful 
  redetermination 
  of 
  

   MgO, 
  several 
  precipitations 
  with 
  NH 
  4 
  HO 
  in 
  presence 
  of 
  suffi- 
  

   cient 
  NH 
  4 
  C1 
  being 
  made. 
  In 
  basic 
  rocks 
  a 
  sodium 
  acetate 
  

   precipitation 
  is 
  always 
  advisable. 
  The 
  results 
  of 
  this, 
  com- 
  

   bined 
  with 
  the 
  reliable 
  determinations 
  of 
  the 
  first 
  analysis, 
  are 
  

   given 
  in 
  III. 
  This 
  new 
  analysis, 
  it 
  will 
  be 
  seen, 
  corresponds 
  

   very 
  closely 
  with 
  that 
  of 
  the 
  rock 
  from 
  La 
  Colonetta, 
  so 
  that 
  

   the 
  chemical 
  character 
  of 
  the 
  type 
  is 
  well 
  fixed. 
  

  

  It 
  is 
  with 
  mixed 
  feelings 
  that 
  I 
  thus 
  repudiate 
  a 
  former 
  

   analysis, 
  but 
  the 
  experience 
  is 
  of 
  service 
  to 
  myself, 
  as 
  well 
  as 
  

  

  * 
  In 
  former 
  papers 
  this 
  name 
  was 
  erroneously 
  given 
  as 
  " 
  Fiesole." 
  following 
  an 
  

   incorrect 
  Austrian 
  map. 
  The 
  spelling 
  ''Fiescoli" 
  is 
  that 
  used 
  on 
  the 
  Italian 
  

   government 
  maps. 
  

  

  + 
  Pirsson, 
  Jour. 
  Geol.. 
  iv. 
  p. 
  688, 
  1806. 
  

  

  JHillebraDd. 
  Bull. 
  148, 
  U. 
  S. 
  G. 
  S., 
  p. 
  39, 
  1897. 
  

  

  >5 
  Washington, 
  Jour. 
  Geol., 
  vii, 
  p. 
  463, 
  1899. 
  

  

  