10     CHEMICAL  COMPOSITION  OP  GELSEMIUM  SEMPERVIRENS. 
water,  from  which  it  is  re-precipitated  by  slight  excess  of  caustic 
potash  and  then  extracted  from  the  mixture  by  ether,  which,  on 
spontaneous  evaporation,  will  leave  it  in  the  form  of  a  very  hard, 
brittle,  transparent  mass,  strongly  adhering  to  the  watch-glass  or 
other  vessel  in  which  the  evaporation  was  effected.  On  care- 
fully detaching  the  residue  and  pulverising  it,  it  will  form  a 
nearly  or  altogether  colorless  powder.  If  the  powder  is  still 
colored,  it  may  be  again  dissolved  and  extracted  by  ether. 
Since  the  alkaloid  is  not  altogether  insoluble  in  water,  a  very 
notable  quantity  will  remain  in  the  filtrate  from  which  the  pre- 
cipitate produced  by  potash  was  separated.  This  may  be  re- 
covered by  precipitating  it  with  a  solution  of  iodine  in  iodide  of 
potassium,  collecting  and  washing  the  precipitate,  then  dissolving 
it  in  alcohol,  and  precipitating  the  iodine  by  the  cautious  addi- 
tion of  nitrate  of  silver,  which  will  throw  it  down  as  iodide  of 
silver,  whilst  the  alkaloid  wiil  remain  in  solution  in  the  form  of 
nitrate.  The  solution  is  then  concentrated  to  expel  the  alcohol, 
diluted  with  water,  filtered,  and  the  filtrate  evaporated  at  a 
moderate  temperature,  when  the  nitrate  will  be  left  in  its  pure 
state.  The  alkaloid  may  readily  be  recovered  from  the  nitrate 
by  dissolving  it  in  water,  adding  slight  excess  of  a  free  alkali, 
and  then  extracting  the  liberated  base  by  ether  or  chloroform. 
Instead  of  employing  the  foregoing  method  for  the  recovery 
of  the  alkaloid  from  the  above  filtrate,  the  liquid  may  be  slightly 
acidulated,  then  concentrated  to  a  small  volume,  again  rendered 
alkaline,  and  the  alkaloid  extracted  directly  by  ether.  To  ob- 
tain it  pure  by  this  method,  however,  will  require  at  least  a 
second  extraction  with  ether. 
In  regard  to  the  proportion  of  the  alkaloid  present  in  the  fluid 
extract  of  gelsemium,  we  obtained,  as  the  average  of  several 
experiments,  about  3-20  grains  of  the  purified  base  from  eight 
fluid  ounces  of  the  extract  examined.  Since  a  fluid  ounce  of 
the  extract  weighs  about  450  grains,  it  would  thus  appear  that 
it  contains  about  1-llOOth  of  its  weight  of  the  alkaloid,  or  about 
one  grain  in  two  and  a  half  fluid  ounces.  Doubtless  a  notable 
quantity  of  the  base  was  lost  in  the  repeated  purifications.  That 
the  extract  as  found  in  commerce  is  uniform  in  strength,  we  are 
not  prepared  to  state. 
