140 
ME.  H.  EMMONS  ON  THE  PETEOGEAPHY  [May  1 893, 
that  mineral.  Experiments  with  hydrochloric  acid  and  hydrate  of 
potash  also  gave  negative  results.  These  facts,  together  with  the 
total  absence  of  crystal-boundaries,  impelled  me  to  consider  it  as 
glass,  which,  in  distinction  from  the  normal  groundmass-glass,  is 
quite  free  from  inclusions. 
Order  of  Crystallization. — So  far  as  the  structure  of  the  andesites 
is  concerned,  there  is  very  little  to  be  said.  With  regard,  however, 
to  the  order  of  crystallization  of  the  various  constituents,  I  could 
find  no  evidence  of  an  4  intratelluric  period.’  Beyond  the  fact  that 
magnetite  and  apatite  are  the  oldest,  the  relative  ages  of  the  rest  of 
the  constituents  can  hardly  be  fixed  with  certainty. 
As  already  remarked,  the  frequency  with  which  inclusions  of 
felspar  occur  in  mica,  and  of  mica  and  augite  in  each  other,  indi¬ 
cates  that  all  three  minerals  must  have  begun  to  separate  about  the 
same  time.  The  felspar  seems  to  have  continued  to  crystallize 
without  appreciable  interruption  throughout  the  entire  period  of 
consolidation  of  the  rock,  as  is  shown  by  the  gradual  diminution, 
without  any  abrupt  break,  in  the  size  of  the  crystals.  In  the  case 
of  mica  and  augite,  however,  there  seems  to  have  been  an  interrup¬ 
tion  in  the  course  of  crystallization,  giving  rise  to  two  generations, 
the  second  of  which  belongs  to  the  groundmass. 
Where  olivine  and  hypersthene  occur,  they  seem  older  than  the 
other  ferro -magnesian  minerals.  So  also  is  the  hornblende  of  the  first 
generation.  With  regard  to  the  hornblende,  it  should  be  remarked 
that  it  never  occurs  in  two  generations  in  the  same  rock  ;  if  it  is 
present  in  the  groundmass,  there  are  no  porphyritic  crystals  of  an 
earlier  age,  and  vice  versa. 
5.  Miceoscopic  Chaeactees  of  the  Anamesites. 
These  are  a  holo  crystalline  variety  of  rock,  composed  of  a  ground- 
mass  of  felspar,  augite,  and  magnetite,  and  porphyritic  crystals  of 
olivine.  Here  and  there  mica-leaflets  may  be  found,  and  a  little 
nepheline.  The  felspar,  which  is  by  far  the  commonest  constituent 
of  the  rock,  occurs  in  lath-shaped  individuals,  and  is  generally  well 
crystallized.  Each  individual  has  two  or  three  t winning-lamellae, 
seldom  more,  parallel  to  its  longest  axis.  The  smallness  of  the 
crystals,  however,  made  it  impossible  to  determine  the  nature  of 
the  felspar  from  its  extinction-angle.  The  augite,  though  next  in 
abundance  to  the  felspar,  is  always  ve^r  small,  and  seldom  shows 
crystal-boundaries.  It  occurs  generally  in  the  shape  of  small 
round  ‘  drops,’  which  occasionally  arrange  themselves,  together 
with  magnetite,  along  the  outer  surfaces  of  felspar-crystals,  or  even 
penetrate  for  small  distances  into  their  mass. 
The  olivine  presents  no  unusual  characteristics.  Nowhere  is  it 
absolutely  fresh.  The  yellow  film  of  iron  oxide  has  already  made 
its  appearance  on  the  surface  and  along  the  cleavage-cracks  of 
every  crystal.  When  crystal-faces  are  observable,  which  is  the 
exception,  they  are  the  domes  and  prisms ;  as  a  rule,  however,  the 
