624 
Extraction  of  Alkaloids  fi^om  Cinchona. 
Am.  Jour.  Pharm.> 
Dec,  1885. 
how  far  it  affects  the  accuracy  of  the  process.  Twenty  grammes  of 
succirubra  bark  were  therefore  treated  in  the  manner  described,  until 
180  cc.  of  percolate  had  been  recovered,  when  caustic  soda  ceased  to 
produce  precipitation.  The  percolation  was  then  resumed  until  further 
950  cc.  had  been  recovered,  and  the  reaction  with  HgI,KaI  had  alsO' 
ceased.  To  this  second  portion  (950  cc.)  a  sufficient  quantity  of  the 
mercuric  reagent  was  added,  and  tlie  liquor  was  allowed  to  settle  for 
some  days,  when  the  clear  supernatant  was  decanted,  and  the  precip- 
itate was  carefully  collected  on  a  filter.  It  was  found  that  while  the 
first  portion  of  the  percolate,  180  cc,  yielded  1*42  gramme  of  alka- 
loids, the  second  portion  yielded  only  0'031  gramme  of  the  compound 
of  alkaloids  with  iodide  of  mercury,  corresponding  to  less  than  0*015 
gramme  alkaloids.  This  inaccuracy,  amounting  to  no  more  than  1 
per  cent,  of  the  alkaloid  present  in  the  bark,  is  so  small  that  it  may,  I 
think,  be  disregarded. 
2.  It  has  been  remarked  that  the  percolate,  after  running  clear  for 
a  considerable  quantity,  presently  becomes  turbid.  I  have  also  some- 
times met  Avith  this  inconvenience,  which  I  attributed  some  years  ago 
to  the  behavior  of  cinchotannic  acid  under  certain  conditions.  In 
analyzing  20  grammes  of  a  sample  of  C.  officinalis,  which  proved  to 
be  very  rich  in  cinchotannic  acid,  I  observed  that  after  97  cc.  of 
])erfectly  clear  percolate  had  been  recovered,  the  succeeding  drops 
caused  a  slight  turbidity  in  the  previously  clear  liquor.  The  percolate 
was  therefore  separated  at  this  point  into  two  portions.  While 
remaining  apart  both  were  perfectly  clear,  but  when  mixed  together  a 
very  turbid  mixture  resulted.  The  explanation  is  simple.  Cincho- 
tannic acid  is  freely  soluble  in  water,  but  scarcely  soluble  at  all  in 
acids.  The  earlier  percolate,  containing  much  hydrochloric  acid, 
dissolved  but  little  cinchotannic  acid;  the  later  percolate,  being  almost 
free  from  acid,  dissolved  much  cinchotannic  acid,  which  was  again 
thrown  out  of  solution  when  brought  into  contact  with  the  previously 
collected  acid  percolate.^  ' 
Although  the  whole  of  the  alkaloids  may  be  extracted  from 
cinchona  bark  either  by  hydrochloric,  phosphoric,  or  nitric  acids,  used 
^  The  above  observations  may  be  further  verified  thus  :  Let  1  ipart  Indian 
cinchona  bark  of  any  species  be  percolated  with  4  parts  water.  To  the 
clear  percolate  add  excess  of  HCl,  and  an  abundant  ijrecipitate  of  cincho- 
tannic acid  will  be  produced.  Ainerican  calisaya  bark,  on  the  contrary, 
shows  no  such  reaction,  in  consequence  of  its  small  amount  of  cinchotannic 
acid. 
