vanhise.1  DISCUSSIONS    OF    PRINCIPLES.  521 
intrude  the  latter.  It  might  be  reasonably  inferred,  if  it  were  not  for 
these  dikes,  that  the  granitic  rock  is  an  eruptive  later  than  the  elas- 
tics (although  the  absence  of  contact  phenomena  would  be  against  this), 
but  as  the  basic  dikes  are  unquestionably  intrusives  of  later  age  than 
the  granites,  and  yet  never  cut  the  slates,  this  explanation  can  not 
possibly  apply.  Evidence  of  this  kind  is  particularly  decisive  if  the 
dikes  are  traced  up  to  the  plane  of  contact  and  have  been  found  to  be 
eroded  or  disintegrated,  as  is  the  case  in  the  Stamford  dike  at  Clarks- 
burg mountain,  Massachusetts,  described  by  Pumpelly,  which  en- 
abled this  author  to  determine  positively  what  had  been  believed  be- 
fore, that  the  granitoid  gneiss  is  unconformably  under  the  Cambrian 
quartzite. 
(5)  Closely  connected  with  (3)  and  (4)  is  degree  of  crystallization  as 
a  guide  to  unconformities.  It  has  been  seen  that  crystalline  character 
is  often  taken  on  in  proportion  as  dynamic  action  occurs.  When  the 
folding,  which  has  affected  only  the  older  series,  has  been  severe,  it  as 
a  whole  will  be  more  crystalline  in  character  than  the  newer.  Also  the 
presence  of  igneous  material  is  often  a  potent  factor  in  the  production 
of  crystalline  character.  As,  however,  recrystallization  is  also  produced 
by  metasomatic  change,  this  criterion  must  be  used  with  caution  and 
as  a  cause  to  search  for  other  evidences  of  an  unconformity  rather  than 
alone  as  a  basis  upon  which  to  infer  an  unconformity.  But  even  differ- 
ence of  amount  of  metasomatic  change,  if  the  rocks  are  equally  likely 
to  be  affected  by  these  processes,  may  be  evidence  of  difference  in  age. 
In  determining  degree  of  crystallization  the  modern  petrographical 
methods  serve  one  of  their  most  useful  purposes,  since  many  rocks 
which  in  exposure  or  in  hand- specimen  appear  to  be  about  equally 
crystalline,  are  shown  in  thin  section  to  be  of  a  fundamentally  different 
character.  A  completely  crystalline  rock  sometimes  can  not  be  dis- 
criminated macroscopically  from  one  which  is  merely  indurated  by 
cementation.  For  instance,  a  thoroughly  crystalline  granite  and  a 
recomposed  rock  built  up  of  the  debris  of  this  granite,  especially  when 
the  particles  are  in  the  form  of  individual  minerals,  rather  than  peb- 
bles, present  much  the  same  appearance  in  mass,  but  a  glance  at  sec- 
tions of  the  two  under  the  microscope  shows  the  thoroughly  crystalline 
interlocking  character  of  the  one  and  the  clastic  character  of  the  other. 
Another  case  quite  as  marked  is  the  discrimination  between  much  foli- 
ated eruptive  rocks  which  have  passed  over  into  fissile  schists  and 
ordinary  argillaceous  slates  and  gr  ay  wackes.  In  the  latter  class  the  par- 
ticles of  quartz  and  feldspar  maybe  seen  with  their  oval  forms  as  regu- 
lar as  the  day  in  which  they  were  deposited,  while  in  the  other  case  an 
entirely  different  appearance  is  presented. 
(6)  Basal  conglomerates  are  one  of  the  most  important  means  of 
determining  a  plane  of  unconformity,  but  it  must  be  clearly  shown  that 
the  conglomerate  is  really  a  basal  one.  Conglomerates  may  occur  in 
other  positions  than  at  basal  horizons,  and  it  will  not  do  to  assume 
