294  Botanical  Source  of  Cuprea  BarJc.        { ^""jll^erim™^ 
Purdie  in  the  fon  sts  of  Antio(^uia,  upon  the  left  bank  of  the  Magda- 
lena  ;  and  Reimjla pcdunculata^  Triana  (Cinchona pedunculatayKixrstcny 
'^Spee.  Select."  !.,  53,  t.  2(j). 
My  identification  of  the  tree  from  the  valley  of  the  Magdalena  river 
is  founded  upon  the  only  samples  that  I  have  received  of  the  cinclion- 
amine-yielding  sort,  which  are  identical  with  those  of  Purdie.  I  incline 
to  believe  that  all  the  other  cuprea  barks  said  to  come  from  Buoara- 
manga,  notwithstanding  the  difference  in  their  chemical  composition 
noticed  by  M.  Arnaud  and  the  not  less  remarkable  difference  in  their 
anatomical  structure  indicated  by  M.  Planchon,  can  only  be  produced 
by  the  same  botanical  species,  viz.,  Remijia  Purdieana ;  inasmuch  as 
(1)  the  barks  containing  cinchonamine  have  been  exported  to  En  rope 
as  being  those  of  cuprea,  without  any  distinction  being  made  between 
them,  except  in  remarking  that  the  trees  from  which  the  bark  was 
obtained  grow  in  a  warmer  locality  at  a  lower  elevation  than  the  others 
without,  however,  indicating  that  they  might  be  different  among  them- 
selves;  and  (2)  if  the  trees  worked  in  the  northern  districts  be  distinct, 
the  resemblance  between  the  one  which  is  most  abundantly  exported 
from  Bucaramanga,  and  which  must  have  been  used  as  a  standard  of 
comparison  to  discover  the  cuprea  bark  in  the  south,  would  be  less 
than  that  which  exists  between  Remijia  Purdieana  and  R.  peduncalata, 
which  is  very  great  at  first  sight. 
The  difference  in  the  conditions  of  vegetation  where  the  trees  yielding 
the  two  kinds  of  cuprea  bark  of  Bucaramanga  grow  would  suffice,  it 
seems  to  me,  to  explain  the  change  in  the  nature  of  the  alkaloids  and 
the  modifications  in  anatomical  structure  observed  in  them.  In  any 
case  this  is  a  question  that  I  hope  I  shall  be  able  to  solve  when  samples 
of  the  common  cuprea  of  Magdalena,  which  I  am  expecting  to  receive 
shortly,  shall  have  arrived.  But  there  can  be  no  doubt  that  if  these 
trees  are  distinct  they  must  belong  to  very  closely  allied  species  of  the 
same  genus. 
With  regard  to  the  southern  district,  I  am  in  possession  of  specimens 
gathered  at  Susumuco,  Villavicencio,  Papamene  and  on  the  banks  of  the 
Guaviare,  etc.,  localities  distant  from  each  other  and  varying  in  eleva- 
tion above  the  sea  level  from  200  to  1,000  meters.  Notwithstanding 
slight  variations,  which  cannot  be  considered  as  specific  characters,  all 
these  specimens  answer  to  Remijia  pedunculata,  Triana,  a  species  dis- 
covered by  M.  Karsten  and  myself  between  Susumuco  and  Villavicencio, 
