28  SMITHSONIAN    MISCELLANEOUS    COLLECTIONS  VOL.    96 
vestigations  properly  form  the  introduction  to  the  work  to  be  described 
in  more  detail  in  the  present  paper,  and  a  preliminary  account  has 
already  been  given  (Collins,  1929).  It  will  be  sufficient  at  this  time  to 
mention  only  the  most  outstanding  results  of  the  Punuk  excavations 
and  to  state  that  the  tentative  chronology  established  on  the  basis  of  the 
Punuk  finds  received  ample  confirmation,  and  elaboration,  through 
the  more  comprehensive  excavations  at  Gambell  in  1930. 
The  kitchen  midden  marking  the  site  of  the  old  village  on  Punuk 
Island  has  a  surface  area  of  400  by  130  feet  and  a  visible  height  of 
about  10  feet,  but  on  digging  through  the  sand  and  gravel  at  the  base, 
we  found  that  refuse  extended  for  another  6  feet  beneath  the  sur- 
face of  the  present  beach  (pi.  2,  fig.  2).  At  the  bottom  were  found 
the  whale  bone  and  timber  remains  of  several  old  house  entrances 
(pi.  2,  fig.  i).  The  fact  that  storm  waves  often  come  right  up  to 
the  foot  of  the  midden  and  6  feet  above  these  old  house  remains 
shows  that  there  has  been  either  a  considerable  subsidence  of  the  shore 
line  since  the  houses  were  abandoned,  or  that  some  outlying  shoals 
or  protecting  reefs  have  been  destroyed,  thus  permitting  the  encroach- 
ment of  the  sea.  On  the  surface  of  the  midden  are  the  pits  of  the 
later  houses,  appearing  now  as  shallow,  leveled  depressions,  with 
nothing  remaining  of  the  superstructures.  Scattered  along  a  flat  sandy 
stretch  to  the  west  of  the  midden  are  the  ruins  of  several  houses  of 
a  considerably  later  period,  the  last  of  which  are  known  to  have  been 
abandoned  about  40  years  ago  (pi.  2,  fig.  3).  Three  of  these  later 
houses  were  excavated,  two  of  them  the  very  latest  and  one  of  them 
probably  dating  from  about  the  beginning  of  the  nineteenth  century. 
The  large  collection  of  artifacts  obtained  from  these  recent  houses 
and  the  shallow  midden  deposits  around  them  provided  a  fairly  com- 
prehensive picture  of  modern  St.  Lawrence  culture  and  afforded  a 
valuable  basis  for  comparison  with  the  much  larger  body  of  material 
excavated  from  the  old  midden. 
When  we  began  to  excavate  at  Punuk,  I  had  hopes  of  finding  further 
evidence  of  the  mysterious  "  fossil  ivory  "  or  Bering  Sea  culture 
which  had  been  discovered  by  Jenness  and  Hrdlicka  in  1926,  particu- 
larly since  most  of  the  beautifully  decorated  objects  of  this  type  which 
Hrdlicka  had  bought  from  the  Eskimos  had  come  from  St.  Lawrence. 
The  work  had  not  progressed  very  far,  however,  before  it  became 
apparent  that  the  Punuk  midden  belonged  to  a  different,  and  presum- 
ably later,  stage  of  culture.  True,  the  artifacts  which  we  were  excavat- 
ing were  for  the  most  part  very  different  from  those  of  the  modern 
St.  Lawrence  Eskimos,  and  the  most  important  group,  the  harpoon 
heads,  resembled  in  form  some  of  those  which  had  been  obtained  by 
