338  SMITHSONIAN    MISCELLANKOUS    COLLECTIONS  VOL.    96 
follows  that  in  America.  It  is  essentially  a  boreal  distribution  and  only 
in  the  marginal  areas  have  the  forms  penetrated  very  far  to  the  south- 
ward— to  New  England  and  southern  British  Columbia  in  America 
and  to  Indo-China  in  Asia.  The  fact  that  slate  blades,  even  though 
later  than  those  of  chipped  stone,  have  been  in  use  among  the  northern 
peoples  since  Neolithic  times  suggests  that  they  are  one  of  the  specific 
elements  of  the  "  coast  "  culture — perhaps  one  of  the  later  elements — 
the  southward  diffusion  of  which  was  limited  either  to  those  coastal 
regions  where,  just  as  in  the  north,  a  maritime  form  of  culture  pre- 
vailed, or  to  interior  regions  like  Manchuria  and  China  to  which 
northern  culture  influences  have  manifestly  extended.  Nowhere,  how- 
ever, has  slate  entirely  supplanted  the  harder  varieties  of  stone;  the 
original  chipping  technique  has  been  retained,  even  by  the  modern 
Eskimos,  though  not  to  the  same  extent  as  by  such  earlier  Eskimo 
cultures  as  those  of  South  Alaska,  the  Old  Bering  Sea,  Dorset,  etc. 
Sledges  and  Toboggans 
One  of  the  most  surprising  results  of  the  Gambell  excavations  is 
the  disclosure  that  dog  traction  seems  to  have  been  unknown  on  St. 
Lawrence  Island  until  about  the  eighteenth  century.  We  should  not 
have  been  unprepared  to  find  that  dog  traction  was  relatively  recent, 
for  scarcely  any  evidence  of  it  had  appeared  at  the  old  sites  around 
Barrow  (Mason,  1930,  p.  385;  Mathiassen,  1930  a,  pp.  84-88).  Fur- 
thermore. Birket-Smith  (1929,  vol.  2.  p.  169)  has  shown  that  it  could 
hardly  have  been  known  to  the  northern  Indians  in  pre-Columbian 
times;  and  Hatt  ( 1934,  pp.  2761-2762),  pointing  out  that  it  is  a  spe- 
cific Eskimo  element,  unknown  originally  from  other  jiarts  of  America, 
feels  that  it  must  therefore  have  reached  America  at  a  comparatively 
late  period.  However,  it  is  surprising  that  dog  traction  should  prove 
to  be  so  very  recent  on  St.  Lawrence  Island.  Mason  thought  that  the 
absence  of  any  evidence  of  dog  traction  at  the  old  1  Harrow  sites  was 
purely  accidental.  Mathiassen  (1930  a,  p.  88)  was  likewise  inclined 
toward  this  view. 
Having  regard  to  the  geographical  distribution  and  great  age  of  dog  traction 
in  Eurasia  and  the  Central  Eskimo  region,  we  must  take  it  as  being  quite  pre- 
cluded that  the  Western  Eskimos  did  not  know  it  in  some  form  or  other  in  olden 
times;  but  it  does  not  seem  to  have  played  any  prominent  part  in  the  culture. 
Dog  traction  is  such  an  imi)ortant  feature  of  Eskimo  life,  that, 
as  Hatt  has  observed,  it  is  difficult  to  imagine  an  Eskimo  culture 
without  it.  And  yet,  if  our  interpretation  is  correct,  we  now  see  that 
for  many  centuries  there  existed  on  St.  Lawrence   Island  and  ap- 
