284  SMITHSONIAN    MISCELLANEOUS    COLLECTIONS  VOL.    96 
preceded  the  circular  ones  in  that  region The  first  white  man  who  came 
into  the  region,  the  explorer  Captain  Cook,  reported  circular  dwellings.  His 
artist  sketched  them,  both  on  the  Aleutian  Islands  and  on  the  mainland  north  of 
the  Strait.  The  square  houses  of  today  seem  to  be  due  to  contact  with  the 
Russians 
If  we  turn  to  Cook's  description  (1784,  p.  108),  however,  we  see 
that  the  only  reference  he  makes  to  the  shape  of  the  Aleutian  houses 
is  that  they  were  "  oblong."  But  by  this  he  seems  to  have  meant 
rectangular,  for  Webber's  illustration  accompanying  the  description 
shows  the  interior  of  an  unmistakably  rectangular  house.  The  other 
Alaskan  dwelling  described  and  illustrated  in  Cook's  narrative  was 
in  Norton  Sound  and  here  again  Webber's  drawing  (opposite  p.  75) 
is  clearly  that  of  a  rectangular  house.  The  only  round  houses  illus- 
trated by  Webber  (opposite  p.  32)  are  some  skin-covered  Chukchee 
houses  on  the  Siberian  side  of  Bering  Strait. 
Birket-Smith  says  (1929,  vol.  2,  p.  53)  :  "Round  houses  with  a 
roof  entrance  are  again  found  on  both  sides  of  the  plank  house :  on 
the  one  hand  among  the  Aleut  (in  an  altered  form),  Koryak  and 
Kamchadal "    However,  as  we  have  seen  above,  the  Aleutian 
and  Kamchadal  houses  are  rectangular  and  the  Koryak  octagonal. 
It  may  be  said,  therefore,  that  no  evidence  has  yet  appeared  to  show 
that  round  houses  have  ever  been  used  in  Alaska,  even  though  on 
theoretical  grounds  these  might  be  expected  to  occur.  As  far  as  I  am 
aware  the  nearest  approach  to  the  round  form  is  found  in  certain 
octagonal  structures  at  Kulukak  and  Goodnews  Bay  in  southwest 
Alaska.  On  the  other  hand  archeological  investigations  on  St.  Law- 
rence Island  have  clearly  demonstrated  the  antiquity,  in  that  area  at 
least,  of  the  square  or  rectangular  house,  an  antiquity  extending  many 
centuries  beyond  the  time  of  Russian  contact.  All  of  the  prehistoric 
St.  Lawrence  houses  were  quadrilateral,  beginning  with  those  of  the 
Old  Bering  Sea  period  and  ending  with  the  latest  prehistoric  form 
which  was  in  use  as  late  as  40  years  ago.  The  only  approach  to  a 
round  house  is  the  octagonal  skin-covered  surface  structure  recently 
adopted  from  the  Siberian  Eskimo. 
The  archeological  data  for  St.  Lawrence  Island  are  of  interest  also 
as  showing  how  little  relation  there  is  between  the  shape  of  a  house 
and  the  material  of  which  it  is  constructed.  Sarfert,  Steensby,  and 
Mathiassen,  in  their  studies  of  Eskimo  habitations,  have  all  assumed 
that  the  shape  of  a  house  is  dictated  by  the  nature  of  the  building  ma- 
terial available,  that  a  whale  bone  house  must  necessarily  be  round 
and  a  wooden  house  quadrangular.  It  is  true  that  walls  made  of  logs 
or  planks  laid  horizontally  can  hardly  be  round,  but  if  the  wall  timbers 
