'^^"'■jan.^'im  Analysis  af  Bitter  Wine  of  Iron,  3 
ting  the  chloroformic  layers  and  mixing  them,  they  were  evaporated 
in  a  tared  beaker  on  a  water-bath,  and  the  residue  dried  at  a  tempera- 
ture of  100°  C.  From  the  weight  of  the  residue  the  percentage  of 
the  alkaloid  was  calculated.  The  residue  from  No.  1  weighed  0.096 
gm.,  equal  to  0.48  per  cent,  of  quinine  present.  The  results  obtained 
from  the  other  samples  proved  but  two  of  them  to  contain  the  offici- 
nal article.  These  two,  Nos.  4  and  5,  were  assayed  with  a  result  of 
0.125  percent,  and  0.45  per  cent,  of  quinine.  'No.  4,  therefore,  being 
only  of  J  the  officinal  percentage,  and  No.  5  lacking  but  0.03  per  cent. 
The  solutions  of  these  alkaloids  in  water  acidulated  with  dilute  sul- 
phuric acid,  had  a  vivid  blue  fluorescence  and  responded  to  the  fol- 
lowing tests  and  reactions  for  quinine  : 
1.  To  a  portion  of  the  solution  fresh  chlorine  water  was  added,  and 
then  NH^OH  in  slight  excess,  when  a  green  color  was  produced  due 
to  thalleioquin. 
2.  To  another  portion  chlorine  water  was  added,  then  a  few  drops 
of  K4Fe(CN)g,  and  finally  a  few  drops  of  NH^OH,  a  red  color  was 
produced,  which  rapidly  faded. 
3.  KOH  added  to  another  portion  formed  a  white  precipitate,  in- 
soluble in  an  excess.  NH^OH  also  produced  a  white  precipitate, 
soluble  in  an  excess. 
The  residues  from  Nos.  2  and  3  for  alkaloid  were  found  to  repre- 
sent respectively  0.4  per  cent,  and  0.225  per  cent.,  the  former  there- 
fore within  0.08  per  cent.,  the  latter  less  than  J  the  officinal  percent- 
age. The  above  mentioned  tests  for  quinine  were  applied  to  acidu- 
lated aqueous  solutions  of  the  suspected  alkaloids,  but  with  negative 
results  in  both  cases ;  these  acid  solutions  also  having  but  an  exceed- 
ingly faint  fluorescence.  The  following  tests  for  cinchonidine  were 
then  applied  to  these  two  solutions,  and  with  positive  results  : 
1.  On  adding  NH^OH  to  some  of  the  solution  a  white  precipitate 
was  formed,  which  was  almost  insoluble  in  an  excess. 
2.  To  another  portion  of  the  solution  KNaC^H^Og  was  added  in 
slight  excess  and  allowed  to  stand  a  short  time ;  to  the  filtrate  was 
added  a  drop  of  NH^OH.  Not  more  than  a  slight  turbidity  resulted, 
showing  the  absence  of  more  than  traces  of  the  other  cinchona  alka- 
loids. 
The  residue  from  No.  6  showed  only  0.05  per  cent.,  about  1-10  the 
percentage  of  No.  1. 
The  above  mentioned  tests  for  quinine  and  cinchonidine,  also  other 
