74  PRE-CAMBRIAN    GEOLOGY   OF    NORTH    AMERICA. 
is  overlain  unconformably  by  the  Carboniferous.  There  is  little 
doubt  that  it  is  Archean,  but  its  reference  to  that  period  is  on  general 
lithological  grounds.  The  second  type  is  of  the  same  composition  as 
the  first,  except  that  biotite  is  substituted  for  muscovite.  It  has  a 
range  from  the  Ombe  Mountains  west  of  Salt  Lake  Desert  to  the 
California  line.  The  third  type  is  like  the  second  except  that  biotite 
and  hornblende  are  found  together.  This  distribution  is  coextensive 
with  that  of  the  second  type.  The  fourth  type  is  the  most  complex 
in  its  penological  features  of  all  the  families  of  granite,  and  con- 
sists of  quartz,  orthoclase,  plagioclase,  often  equal  in  quantity  to  the 
orthoclase  and  sometimes  exceeding  it,  usually  a  high  percentage  of 
biotite  and  hornblende,  titanite,  and  a  high  proportion  of  micro- 
scopic apatite.  Between  this  class  and  the  diorites  that  are  unusually 
rich  in  orthoclase  there  is  but  little  difference,  although  there  is  little 
danger  of  ever  confounding  the  granitoid  diorite  with  the  dioritic 
members  of  the  fourth  type.  These  granites  are  the  most  prominent 
as  regards  geographical  distribution  of  the  truly  eruptive  varieties 
observed  in  the  Cordilleras.  When  the  different  types  of  granite 
are  seen  in  apposition,  so  as  to  give  a  clue  to  their  relative  ages,  it  is 
found  that  they  occur  in  the  order  given.  In  denominating  these 
groups  of  granite  as  eruptive  it  is  only  intended  to  indicate  that  in 
their  relations  to  the  contiguous  Archean  schists  they  have  the  ap- 
pearance of  intrusive  bodies,  and  that  in  their  interior  structure  and 
general  mode  of  occurrence  there  are  none  of  those  evidences  of 
alliance  to  the  crystalline  schists  which  are  observed  in  the  grani- 
toid gneisses  of  so  many  localities,  especially  in  the  Rocky  Mountain 
region.  In  so-called  eruptive  granites  there  is  parallelism  neither 
of  general  bedding  nor  of  interior  arrangement  of  the  minerals,  and 
the  most  ordinary  phenomenon  of  structure  is  the  development  of 
conoidal  shapes  formed  of  concentric  layers  varying  in  thickness 
from  a  few  inches  to  100  feet.  This  structure,  so  far  as  observed,  is 
strictly  confined  to  the  hornblende-bearing  granites,  and  never  makes 
its  appearance  in  those  of  the  first  and  second  types.  Although 
instances  of  each  type  of  granitic  rock  are  found  unconformably 
underlying  the  low  members  of  the  Paleozoic  series,  this  is  not  the 
case  with  each  outcrop;  many  granitic  masses  are  found  unconform- 
ably underlying  Mesozoic  or  even  Tertiary  volcanic  rocks.  But  there 
is  absolutely  no  evidence  whatever  in  favor  of  the  belief  of  granitic 
extrusions  later  than  the  Archean  age.  Although  Whitney  has  found 
intrusions  of  granite  in  sedimentary  strata  other  than  the  Archean 
crystalline  schists,  any  attempt  to  correlate  age  by  petrological  fea- 
tures alone  is  dangerous,  as  may  be  shown  by  the  fact  that  the  Juras- 
sic granite  of  California  and  the  granite  of  the  Cottonwood  region 
on  the  Wasatch,  which  is  unmistakably  Archean,  are  positively  iden- 
tical down  to  the  minutest  microscopical  peculiarity. 
