30  SMITHSONIAN    MISCELLANEOUS    COLLECTIONS  VOL.    96 
extent.  What  was  the  orig^in  of  this  metal?  If  it  had  been  obtained 
directly  from  the  Russians  after  the  discovery  of  Alaska  in  the  eight- 
eenth century  or  indirectly  through  the  Chukchee  after  the  arrival 
of  the  Cossacks  in  northeastern  Siberia  a  hundred  years  earlier,  it 
would  mean  that  the  Punuk  Island  and  Cape  Kialegak  settlements  were 
less  than  300  years  old.  However,  everything  pointed  to  this  being 
nearer  the  time  of  the  abandonment  of  the  Punuk  midden  than  of  its 
establishment.  Most  of  the  objects  from  the  Punuk  midden  differed 
from  those  found  in  the  recent  house  ruins  on  the  adjoining  sand  flat 
just  as  strikingly  as  the  shallow,  leveled  pits  of  the  older  houses  on 
the  midden  themselves  differed  from  these  later  ruins,  which  even 
though  abandoned  for  40  years,  still  had  the  walls  and  roofs  prac- 
tically intact.  Furthermore,  it  would  seem  incredible  that  such  huge 
kitchen  middens  as  those  at  Punuk  and  Kialegak,  16  and  18  feet  deep, 
had  been  formed  in  less  than  300  years ;  or  that  the  houses  at  the 
bottom  of  the  Punuk  midden,  now  6  feet  beneath  the  reach  of  storm 
waves,  could  have  been  built,  occupied,  abandoned,  and  then  covered 
over  by  an  accumulation  of  debris  16  feet  deep  all  within  so  short  a 
space  of  time.  If  these  earliest  Punuk  Eskimos  possessed  metal,  as 
the  evidence  of  their  art  so  clearly  indicated,  they  must  have  received 
it  long  before  the  arrival  of  the  Russians  in  northeastern  Siberia  in 
the  seventeenth  century.  Its  source  must  for  the  present  remain  in 
doubt,  but  the  fact  that  Laufer  has  shown  that  iron  was  being  used 
in  eastern  Siberia  as  early  as  the  third  century  A.  D.  indicates  at 
least  the  possibility  of  its  having  been  introduced  on  St.  Lawrence 
Island  long  before  the  seventeenth  century. 
In  1929  I  returned  to  Cape  Kialegak,  accompanied  by  G.  Herman 
Brandt,  and  excavated  for  one  month.  While  the  results  were  grati- 
fying in  that  another  large  collection  of  artifacts  was  obtained  which 
increased  our  knowledge  of  the  Punuk  culture,  it  was  again  a  disap- 
pointment that  nothing  was  found  that  threw  any  further  light  upon 
the  elusive  Old  Bering  Sea  culture.  For  just  as  in  the  previous  year, 
Punuk  art  was  found  almost  exclusively,  with  just  enough  Old  Bering- 
Sea  pieces  (9,  to  be  exact — and  again  from  the  lower  levels  of  the 
middens)  to  show  that  it  would  be  necessary  to  find  some  still  older 
site  if  the  Old  Bering  Sea  culture  were  to  be  revealed  in  its  entirety. 
Having  tested  the  ix»ssibilities  of  the  two  largest  and  most  promising 
old  sites  on  the  eastern  end  of  St.  Lawrence,  it  now  seemed  advisable 
to  look  elsewhere,"  and  the  most  promising  place  seemed  to  be  Gambell 
'  I  have  since  learned  of  a  very  important  site  on  Punuk  Island — apparently 
a  pure  Old  Bering  Sea  site — which  completely  escaped  my  attention  when  I 
was  on  the  island  in  1928.    This  site  was  discovered  by  Otto  Wm.  Geist  in 
