332  SMITHSONIAN    MISCELLANEOUS    COLLECTIONS  VOL.    96 
and  Iranian  tribes  of  central  and  western  Asia.  The  evidence  seems 
clearly  to  indicate  that  plate  armor  originated  among  the  warlike 
tribes  of  central  Asia  whence  it  spread  southward  and  eastward  into 
China  and  gradually  northward  toward  Bering  Sea.  It  seems  incredible 
that  it  should  have  penetrated  to  far  northeastern  Siberia  and  been 
adopted  by  the  primitive  hunting  and  fishing  tribes  around  Bering 
Strait  before  it  was  adopted  in  China.  Though  not  by  nature  a  war- 
like people,  the  Chinese  have  always  been  faced  with  the  necessity 
of  protecting  their  lands  from  invasion  and  more  than  once  have 
found  it  expedient  to  adapt  their  own  mode  of  warfare  to  that  of 
their  enemies.  China,  with  its  fertile  valleys  and  rich  civilization,  was 
a  constant  temptation  to  the  restless  barbarian  hordes,  and  the  effects 
of  repeated  foreign  invasion  must  have  been  felt  there  centuries  earlier 
than  in  the  barren  regions  to  the  northward.  Indeed,  it  would  seem 
reasonable  to  suppose  that  the  primitive  tribes  around  Bering  Strait 
would  be  the  last  to  have  been  affected,  even  indirectly,  by  such  move- 
ments, in  which  case  the  seventh  or  eighth  centuries  A.D.,  or  later, 
would  appear  as  a  more  probable  date  for  the  introduction  of  plate 
armor  than  the  third  century. 
The  conclusion  seems  unmistakable  that  plate  armor  whether  of 
bone,  ivory,  or  iron,  is  peripheral  at  Bering  Strait,  the  center  lying 
somewhere  to  the  southward  in  Asia.  This  being  the  case,  how  are 
we  to  explain  the  occurrence  of  armor  elsewhere  in  America?  Since 
wooden  slat  armor,  the  American  type  closest  to  the  plate  armor 
of  the  north,  has  its  center  in  the  North  Pacific  area,  it  would  seem 
plausible  to  assume,  as  has  generally  been  done,  that  the  two  types 
were  related  in  origin.  There  is  some  question,  however,  whether  this 
is  actually  true.  Technically  there  is  as  great  a  discontinuity  between 
Eskimo  and  Northwest  Coast  armor  as  there  is  a  continuity  between 
the  former  and  that  which  is  found  widespread  in  Asia.  The  manner 
of  lashing  together  the  individual  rectangular  plates  and  the  imbricated 
arrangement  of  these  in  parallel  and  somewhat  overlapping  rows  are 
features  which  are  exhibited  in  practically  identical  form  from  Bering 
Strait  to  Central  and  Eastern  Asia,  but  which  are  lacking  completely 
in  America  south  of  Bering  Strait.  If,  in  spite  of  what  seems  to 
have  been  a  fundamental  difference  in  construction,  we  are  still  to 
regard  the  slat  armor  of  the  Pacific  region  as  derived  from  the  plate 
armor  of  Bering  Strait,  we  are  faced  with  the  further  difficulty  that 
the  latter  type  does  not  appear  until  the  Punuk  stage.  Although 
conceivable,  it  would  seem  rather  unlikely  that  the  history  of  armor 
in  America  is  to  be  encompassed  in  so  short  a  period,  i)articularly  if 
the  rod  and  slat  armor  of  the  Atlantic  seaboard  was  related  to  that 
