346  Strophanthiis  Seed  of  Commerce.  {^j&lvSK!"*' 
The  relative  size  of  the  seeds  cannot  well  be  expressed  in  measure- 
ments of  length  and  breadth,  since  in  some  varieties  the  seeds  taper 
more  than  in  others.  The  measurements  of  the  awn  and  plume  are 
those  of  seeds  taken  from  the  middle  of  the  follicle,  and  the  meas- 
urements of  the  length  of  the  follicles  in  the  Society's  Museum  are 
taken  from  those  containing  mature  seeds,  the  sign  -f  indicating 
that  the  follicle  is  broken  at  the  apex  and  is  probably  longer  when 
perfect.  The  breadth  is  not  given,  since  all  the  follicles  are  open 
in  different  degrees,  and  a  comparison  would  not  be  of  practical  value. 
No.  I  is  undoubtedly  the  seed  of  the  plant  described  as  5.  hispidus, 
and  Nos.  2,  3,  4,  5  are  probably  derived  from  forms  of  the  same 
species. 
No.  6  was  received  from  Dr.  J.  F.  Easmon,  and  the  only  herbarium 
specimens  of  Strophanthiis  received  from  him  appears  to  be  5. 
Preussii,  but  he  does  not  state  that  the  seeds  and  fruit  were  derived 
from  the  same  plant. 
No.  8  was  received  from  Mr.  G.  F.  Scott  Elliott,  who  observed 
only  two  species  in  Sierra  Leone,  hispidus  and  ^S.  sarinentosus,  and 
he  believes  that  the  fruit  was  probably  gathered  from  the  latter, 
since  the  seeds  give  a  different  reaction  to  those  from  5".  hispidus. 
With  No.  1 1  some  leaves  were  received  as  those  of  the  plant 
yielding  the  seed,  and  these  leaves  possess  the  curiously  punctate 
appearance  of  those  of  S.  gracilis.  No.  9  was  no  doubt  obtained 
from  5.  hispidus,  var.  Kombe,  the  plant  figured  by  Professor  Fraser. 
No  clue  can  be  obtained  as  to  the  botanical  source  of  the  other 
specimens. 
It  should  be  observed  that,  in  applying  the  sulphuric  acid  test, 
allowance  must  be  made  for  the  fact  that  in  some  seeds  the  color 
is  developed  much  more  slowly  than  in  others,  and  in  a  slightly 
different  manner  in  different  varieties.  Thus  in  5.  Kombe  the 
endosperm  becomes  dark  green  almost  immediately,  and  this  color 
usually  spreads  over  one  of  the  cotyledons,  the  other  becoming 
purplish.  In  S.  hispidus,  the  green  color  does  not  appear  until  after 
the  lapse  of  a  minute  or  longer,  and  does  not  usually  spread  to  the 
cotyledons.  The  purplish  tint  of  the  cotyledons  does  not  readily 
darken  as  it  does  in  S.  Kombe.  In  some  apparent  varieties  of  S.  hispi- 
dus, the  color  of  the  cotyledons  becomes  of  a  bluish  rather  than  a 
reddish  purple.  In  the  species  containing  no  strophanthin  the  seeds 
vary  to  a  certain  extent,  also,  in  the  rapidity  with  which  the  red  color 
