vanh.se.]  DISCUSSIONS    OF    PRINCIPLES.  481 
the  clastic  rocks  has  the  objection  that  it  is  an  unverified  hypothesis. 
When  once  a  sedimentary  rock  has  become  fluent  and  is  wholly  flee  to 
crystallize  anew  how  shall  the  material  be  identified?  To  state  that 
such  material  has  not  moved  far  is  pure  assumption.  If  the  fusion 
theory  is  true  the  average  composition  of  the  unfused  part  of  the  clas- 
tic series  and  the  subjacent  material  should  agree.  To  obtain  average 
analyses  of  rocks  which,  vary  widely  in  mineral  character  within  short 
distances  is  not  easy,  but  is  a  thing  which  must  apparently  be  attempted 
if  this  theory  is  to  be  maintained,  for  the  writer  sees  no  other  way  in 
which  an  attempt  can  be  made  to  verify  the  hypothesis. 
Another  class  of  geologists,  noting  these  contact  relations  between 
the  granitic  rocks  and  the  elastics,  hold  that  the  former,  called  by  others 
Lower  Laurentian  or  Archean,  are  eruptives  of  later  age  than  the 
elastics  with  which  they  are  in  contact,  without  attempting  to  give  any 
theory  as  to  the  source  of  the  material.  Here  are  included  Hawes, 
Hall,  Mather,  Foster,  Whitney,  W^adsworth,  Rominger,  Herrick,  and 
others.  Eominger  distinctly  recognized  the  granites  and  granite- 
gneisses  of  this  kind  on  the  south  shore  of  lake  Superior  as  the  subjacent 
rocks  upon  which  the  schists  rest.  Herrick  saw  the  same  relations  with 
reference  to  his  granitic  and  schistose  groups  on  the  north  shore.  In  the 
last  two  cases  the  facts  before  the  writers  are  precisely  the  same  as  those 
of  the  geologists  of  the  second  school ;  but,  by  giving  no  explanation  of 
the  source  of  the  material  for  the  granite  gneisses,  they  have  esoaped 
the  difficulty  of  the  unverified  assumption  that  these  eruptives  repre- 
sent fused  sediments.  They  fail  to  explain  what  has  become  of  the 
floor  upon  which  the  elastics  are  deposited.  Some  floor  they  must  have 
had.  Where  these  eruptive  contacts  are  found  the  floor  has  disa£>- 
peared,  and  if  so  the  eruptives,  if  extraneous,  must  be  considered  to 
have  eaten  up  or  absorbed  it. 
The  three  positions,  that  the  granite-gneiss  is  selectively  metamor- 
phosed material,  is  due  to  subcrustal  fusion,  or  is  an  extraneous  intru- 
sive, may  be  considered  to  grade  into  each  other.  Those  who  insist  on 
the  first  have  selective  metamorphism  and  contact  phenomena  to  ex- 
plain. Those  who  insist  on  subcrustal  fusion  may  be  called  upon  to 
identify  their  material.  They  can  only  show  the  former  fused  condi- 
tion by  contact  phenomena,  and  contact  phenomena  are  not  evidence 
of  progressive  fusion,  but  intrusion.  Those  who  regard  the  granite- 
gneiss  as  intrusives  may  be  asked  what  has  become  of  the  floor  upon 
which  the  elastics  were  deposited. 
(3)  That  the  Archean  is  an  igneous  system  earlier  than  any  of  the  sedi- 
inentaries  is  apparently  the  conclusion  of  Emmons,  Lieber,  and  others. 
These  careful  observers  not  only  maintained  the  igneous  origin  of  the 
granite-gneiss  of  the  southern  Apx>alachians,  but  traced  the  gradations 
between  basic  schistose  rocks  and  massive  eruptives,  including  horn- 
blende schist  and  unmistakable  dikes,  and  drew  the  correct  conclusion, 
lately  regarded  as  a  new  discovery,  that  such  rocks  are  sometimes 
Bull.  8(5 31 
