66  PRE-OAMBRIAN    ROCKS    OF    NORTH    AMERICA.  [bull. 86. 
second  is  at  Grassy  lake,  where  there  is  a  pebbly  conglomerate  at  the* 
base  of  the  Keewatin.  The  contacts  of  the  Keewatin  and  Coutchi- 
ching are  usually  in  apparent  conformity,  and  the  mapping  of  the  series 
is  made  to  rest  upon  lithological  characters  rather  than  upon  structural 
discordance,  although  it  is  recognized  that  the  conglomerates  at  the 
two  localities  mentioned  are  indicative  of  erosion,  at  least  in  some 
places,  between  the  two  series.  The  apparent  accordance  affords  little 
evidence  as  to  the  question  of  original  conformity  or  unconformity, 
because  the  two  formations  have  been  squeezed  together;  but  the 
marked  contrast  in  lithological  characters  indicates  an  abrupt  change 
in  the  conditions  of  formation. 
The  thickness  of  the  Keewatin  rocks  is  calculated  to  be  about  live 
miles.  The  rocks  are  found  in  the  main  to  be  of  clastic  origin,  but 
probably  in  the  nature  of  volcanic  debris  rather  than  water- deposited! 
sediments,  although  a  small  quantity  of  the  materials  is  of  the  latter 
character.  The  volcanic  debris  of  the  Keewatin  has  a  basic  and  an: 
acid  division,  the  latter  being  higher  in  the  series  than  the  former. 
Microscopical  study  leads  to  the  conclusion  that  the  series  has  been 
subjected  to  great  i>ressure,  as  is  evidenced  by  the  fracture  of  the* 
grains  as  well  as  by  the  schistose  structure.  The  granite  •  bowlders- 
found  in  the  conglomerates  may  either  be  ordinary  detritus  or  may\ 
have  been  brought  up  from  beneath  by  volcanic  forces,  although  the 
former  is  the  more  probable.  All  of  the  granites  of  the  region  appear 
to  be  in  some  degree  later  than  the  Keewatin  rocks,  but  this  does  not 
imply  that  there  was  not  a  granite  shore  for  the  basin  in  which  the 
Keewatin  rocks  were  deposited.  The  author  has  no  doubt  that  the 
original  floor  upon  which  the  Keewatin  and  Coutchiching  rocks  were] 
deposited  was  fused  at  the  time  of  disturbance  and  appears  now  to  us 
as  the  foliated  granite  of  the  Laurentian.  The  granite  bowlders  of  the 
Keewatin  agglomerates  may  have  been  derived  from  a  granite  basement 
now  obliterated  by  subsequent  plutonic  fusion. 
The  Coutchiching  series  consists  of  mica-schists,  or  mica- schists  in 
which  feldspar  is  present.  Hornblende  has  been  observed  only  in  one 
instance.  Ep  limestones  or  conglomerates  have  been  discovered.  Thee 
schistose  structure  is  believed  to  represent  original  sedimentation.  Cal-j 
dilating  the  thickness  upon  this  basis,  it  is  found  that  the  formation 
has  an  average  thickness  of  4  or  5  miles.  The  relations  of  the  Coutchi-^ 
ching  series  to  the  Laurentian  are  found  to  be  precisely  the  same  as 
between  the  Keewatin  and  Laurentian.  In  the  Coutchiching  series 
there  are  no  intercalations  of  recognizable  volcanic  rocks  such  as  arq 
found  in  the  Keewatin,  the  rocks  being  acid  crystalline  schists  which 
are  regarded  as  ordinary  metamorphosed  quartzose  sediments,  with 
perhaps  also  volcanic  material,  although  in  no  place  has  it  been  possi- 
ble conclusively  to  prove  this. 
The  Laurentian  gneiss  is  intermixed  with  granite  in  such  a  way  as  to 
make  it  impossible  at  times  to  separate  them.     Its  structure,  if  it  hasi 
