NO.    I  ARCHi:0LOGV    OF    ST.    LAWRENCE    ISLAND COLLINS  3II 
We  find  a  similar  condition  with  regard  to  the  harbed  form,  Thule 
type  2.  Here  again  the  Gambell  finds  throw  no  light  on  the  origin  of 
the  type,  for  it  appears  quite  suddenly  and  rather  late  in  the  sequence, 
at  levoghiyoq  (pi.  70,  figs.  4-6 — open  socket  type  IV).  It  is,  however, 
earlier  at  Gambell  than  the  Thule  type  i,  which  appeared  only  at  Sek- 
lowaghyaget  and  the  old  section  of  Gambell.  The  present  evidence 
indicates  that  the  Thule  type  2  head  originated  on  the  Arctic  coast, 
probably  around  Point  Barrow,  and  was  later  introduced  on  St. 
Lawrence  Island ;  for  at  Barrow  the  barbed  form  holds  an  important 
place,  from  the  "  Birnirk  "  stage,  when  there  was  usually  one  promi- 
nent barb  with  an  opposite  side  blade,  to  late  prehistoric  times  when 
the  form  with  two  or  even  four  barbs,  but  with  a  closed  socket, 
became  common  (Collins,  1935,  pi.  ii,  figs.  3,  4). 
Because  of  its  relatively  simple  form  and  wide  distribution,  Mathias- 
sen  considers  that  the  Thule  type  2  head  is  older  at  Point  Barrow  than 
the  Birnirk  type,  which  was  the  prevailing  form  revealed  by  the 
excavations  of  Stefansson  and  Van  Valin  at  old  abandoned  sites 
around  Barrow : 
....  the  Birnirk  types  are  almost  predominant  in  the  Van  Valin  collection, 
which  otherwise  contains  hardly  anything  else  than  implements  of  the  Thule 
types,  and  also  in  the  Birnirk  collection,  which  forms  a  transition  from  the  Van 
Valin  stage  to  the  later  Pt.  Barrow  culture,  whereas  the  Thule  harpoon  heads 
do  not  occur  in  any  of  these  collections  and  thus  must  presumably  be  earlier 
than  these  ....  judging  by  their  shape  the  Birnirk  types  seem  to  have  been 
derived  from  the  Thule  harpoon  heads  and  presumably  have  appeared  as  the 
result  of  the  influence  of  the  Thule  types  from  the  Bering  Sea  Culture  .... 
this  influence  must  have  been  brought  to  bear  after  the  Thule  Culture  had 
spread  from  Alaska  to  the  east.    [Mathiassen,  1930  a,  p.  33.] 
It  is  difficult  to  see  why  the  absence  of  the  Thule  heads  in  these 
archeological  collections  should  be  taken  as  presumptive  evidence  of 
their  having  belonged  to  some  still  earlier  period,  particularly  since 
these  same  Thule  heads  were  by  no  means  lacking  in  the  collections 
from  northern  Alaska  which  were  purchased  from  the  Eskimos.  A 
more  likely  explanation  would  be  that  the  collections  obtained  from 
the  Eskimos  were  relatively  recent  ( for  wherever  the  Alaskan  Eskitnos 
dig  for  specimens  they  choose  the  most  accessible  ruins,  and  these 
are  usually  the  most  recent)  whereas  the  material  excavated  by  Stef- 
ansson and  Van  Valin  at  old  "  mound  "  sites  away  from  the  present 
villages  represented  an  earlier  stage  of  culture. 
Confirmation  of  this  view  seems  to  have  been  furnished  by  the  more 
recent  systematic  excavations  made  at  these  same  Barrow  sites  by 
James  A.  Ford  for  the  Smithsonian  Institution,  the  principal  results 
of  which  have  been  summarized  by  the  writer  (Collins,  1933,  pp.  45- 
