THE   CORDILLERAS.  797 
ings  in  the  strata  is  it  safe  to  refer  it  to  a  later  age  than  the  sedimen- 
tary series. 
In  the  analytical  map  the  rocks  are  divided  into  two  classes,  the  in- 
tention being  to  discriminate  those  formations  which  are  sedimentary 
from  the  class  of  eruptive  rocks ;  but  this  line  can  not  be  drawn  with 
precision,  because  the  series  of  gneisses  pass  into  the  massive  layers 
and  because  limited  bodies  of  granite  which  are  massive  might  if 
more  largely  exposed  pass  into  crystalline  schists  or  other  Archean 
sedimentary  rocks. 
It  is  not  easy  to  analyze  those  subtle  appearances  which  lead  the 
observer  to  incline  to  one  or  the  other  of  the  two  possible  modes  of 
origin  of  a  granite  outcrop.  Parallelism  of  bedding,  and  even  paral- 
lelism of  the  arrangement  of  minerals,  are  consistent  with  the  theory 
of  an  eruptive  origin.  Certain  masses  of  gneissoid  granite  appearing 
in  the  great  eruptive  granite  body  of  the  Sierra  Nevada  show  quite 
as  much  parallelism  of  bedding  and  internal  arrangement  of  minerals 
as  the  Rocky  Mountain  granites  to  which  Ave  have  assigned  a  meta- 
morphic  origin;  yet  the  Sierra  field,  as  a  whole,  is  clearly  eruptive. 
But  at  the  same  time,  in  the  intimate  arrangement  of  the  mineral 
particles  and  in  the  mode  of  contact  between  the  various  mineral 
ingredients,  there  is  a  certain  broad  uniformity  in  all  the  eruptive 
granites,  which  produces  a  characteristic  impression  upon  the  eye. 
On  the  contrary,  the  granites  which  Ave  conceiATe  to  have  been  of 
metamorphic  origin,  no  matter  Iioav  simple  the  mineralogical  compo- 
sition, have  always  a  peculiar  variability  of  arrangement;  and  even 
in  the  absence  of  any  pronounced  parallelism  they  show  the  effect  of 
interior  compression  and  irregular  mechanical  influences.  In  the 
eruptiAre  granites  there  seems  to  have  been  a  steady  expansiA^e  force, 
doubtless  due  to  the  heat  and  elastic  fluids,  which  gave  to  all  the 
particles  a  certain  independent  polarity,  while'in  the  metamorphic 
granites  they  seem  to  have  been  crowded  into  constantly  conflicting 
positions.  As  the  result  of  this,  the  crystalline  particles  of  the  meta- 
morphic granites  are  much  less  apt  to  have  completed  their  crystal- 
lization, or,  if  it  was  completed,  they  haA^e  been  crushed  and  torn 
asunder  and  their  particles  scattered,  Avhile  in  the  case  of  the  eruptive 
granites  crystallization  seems  to  have  been  more  perfected.  The  re- 
sult of  this  is  to  give  to  the  eruptive  granites  something  of  the  uni- 
formity of  texture  of  a  volcanic  rock,  Avhile  all  the  metamorphic 
granitoid  rocks,  when  once  the  gneissoid  parallelism  of  minerals  is 
broken  up,  have  a  crushed,  irregular,  and  confused  mode  of  arrange- 
ment. 
The  metamorphic  rocks  of  the  Humboldt  Mountains,  Franklin 
Buttes,  and  the  Kinsley  district  are  provisionally  correlated  with  the 
Huronian  of  Canada. 
