678  PRE-CAMBRIAN    GEOLOGY   OF    NORTH   AMERICA. 
The  Transition  argillite  is  widespread,  and  in  it  occur  most  of  the 
gold  mines. 
Mitchell,65  in  1842,  describes  as  Primitive  formations  the  granites, 
gneiss,  mica  slate,  chlorite  slate,  hornblende  slate,  and  talcose  slate, 
quartz  rock,  serpentine,  and  limestone.  A  vast  body  of  granite  trav- 
erses the  State  in  a  northeast-southwest  direction,  comprising  a 
large  part  of  Person,  Caswell,  Orange,  Guilford,  Randolph,  Davidson, 
Rowan,  Cabarrus,  and  Mecklenburg  counties;  also  some  of  Lincoln, 
Iredell,  Davie,  Stokes,  and  Rockingham  counties.  Within  this  belt  is 
no  well-defined  gneiss,  micaceous  Primitive  slate,  serpentine,  or  lime- 
stone. West  of  this  formation  are  the  most  ancient  Primitive  rocks, 
on  the  upper  waters  of  the  Dan,  Yadkin,  Catawba,  and  French 
Broad.  Here  are  a  great  variety  of  granites.  Gneiss  and  slate  also 
occur.  All  of  these  are  interstratified.  Limestones  are  found  at  three 
points  in  Stokes  County.  In  Anson  and  Richmond  counties  is  a 
beautiful  porphyritic  granite.  East  of  the  red  sandstone  in  the 
counties  of  Cumberland,  Wake,  Granville,  Warren,  Franklin,  Nash, 
Johnston,  Halifax,  and  Northampton  is  another  body  of  ancient 
Primitive  rock  in  which  granite  prevails. 
Emmons  (E.),66  in  1856,  gives  a  systematic  account  of  the  crystal- 
line rocks  of  North  Carolina.  Rocks  of  igneous  origin  are  often  mas- 
sive, but  also  frequently  are  laminated,  and  laminated  rocks  are 
frequently  called  stratified,  but  this  term  should  be  restricted  to  the 
sedimentary  rocks.  The  metamorphic  rocks  are  excluded  from  the 
sedimentary  classification  because  all  rocks  may  become  metamorphic, 
and  a  stratum  metamorphic  in  one  locality  may  not  be  metamorphic 
in  another.  The  highest  proof  of  the  age  of  rocks  is  the  order  of 
superposition.  When  this  method  can  be  applied  it  is  paramount,  but 
paleontology  may  be  used  subject  to  proper  principles.  At  the  base 
of  the  Paleozoic,  under  the  Silurian,  is  placed  the  Taconic. 
The  granitic  formations  are  regarded  as  eruptive  or  pyrocrystal- 
line.  They  form  two  continuous  belts,  which  cross  the  State  in  a 
northeast-southwest  direction.  The  eastern  one  is  the  Raleigh  belt, 
and  the  western  one  the  Salisbury  and  Greensboro  belt.  Granite  is 
generally  the  underlying  rock,  but  there  are  cases  on  record  in  which 
it  is  shown  that  it  is  an  overlying  one.  At  Warrenton,  in  Warren 
County,  of  the  Raleigh  belt,  it  is  found  to  overlie  gneiss,  mica  slate, 
and  hornblende,  where  it  is  considered  to  have  been  projected  through 
fissures  in  these  rocks.  This  eastern  belt  contains  no  metallic  veins, 
nor  is  it  cut  by  trap  or  other  intrusive  rocks.  Its  breadth  is  from 
20  to  25  miles. 
The  Salisbury  granite  is  frequently  syenitic — that  is,  hornblende 
takes  the  place  of  mica.  This  belt  is  cut  by  numerous  peculiar  dike 
rocks,  in  which,  when  they  decompose,  the  hornblende  trap  appears  in 
dark-green  stripes,  and  many,  when  carefully  examined,  have  as- 
