38  Estimation  of  Berberine  %  {A?anwyfX.m' 
berine  is  extracted  from  the  drug  by  means  of  hot  alcohol.  In  the 
first  of  these  methods  the  alcoholic  solution  is  made  up  to  a  definite 
volume,  filtered  and  in  an  aliquot  portion  of  the  alcoholic  filtrate 
the  berberine  precipitated  as  an  acid  sulphate  which  is  afterwards 
converted  into  the  monoacid  hydriodide.  In  applying  this  method 
to  various  drugs  it  was  found  that  whereas  in  some  of  them,  like 
Hydrastis  canadensis,  the  berberine  exists  in  a  form  that  is  easily 
soluble  in  alcohol ;  in  others,  like  some  samples  of  barberry  bark, 
the  alkaloid  seems  to  exist  in  a  form  which  is  very  difficultly 
soluble  in  cold  alcohol,  so  that  it  is  impossible  to  make  up  the  alco- 
holic extract  with  cold  alcohol  to  a  definite  volume  without  leaving 
some  berberine  in  the  residue.  It  is  therefore  important:,  when- 
ever the  first  method  of  estimation  is  used,  to  take  care  that  no 
berberine  containing  residue  be  left. 
Should  there  be  a  sediment  in  the  alcoholic  extract,  which  does 
not  dissolve  in  cold  alcohol,  even  after  repeated  washing  with  this 
solvent,  the  sediment  should  be  dissolved  in  water,  filtered  with  a 
little  talcum,  if  necessary,  and  the  clear  filtrate  tested  for  berberine 
by  one  of  the  methods  which  will  be  given  in  the  next  paper.  If 
berberine  be  found  in  the  sediment  then  the  first  method  of  estima- 
tion is  inapplicable,  and  the  second  method,  which  is  applicable  to 
all  cases,  must  be  used. 
With  regard  to  the  precipitation  of  berberine  by  potassium 
iodide,  further  work  upon  this  subject  has  shown  that,  whereas  in  a 
neutral  solution  berberine  is  as  completely  precipitated  by  an 
excess  of  potassium  iodide,  as  it  is  by  Mayer's  or  Wagner's  reagent, 
in  the  presence  of  free  acid  the  precipitation  by  these  latter  reagents 
is  more  complete  than  by  potassium  iodide.  It  is  therefore  best  to 
substitute  Mayer's  reagent  as  a  precipitant  for  potassium  iodide  in 
the  above-mentioned  first  method,  in  which  acid  is  set  free  by  the 
reaction. 
That  the  reaction  is  the  same  whatever  neutral  precipitant  be 
used,  was  shown  in  another  paper,1  where,  on  adding  standard  acid 
to  a  neutral  solution  of  berberine  hydrochloride  and  precipitating 
the  alkaloid  with  either  of  the  above-mentioned  reagents,  no  acid 
was  consumed,  showing  that  even  in  the  presence  of  free  acid,  ber- 
1  Arch.  d.  Pharm.,  1901,  629.  Read  at  St.  Louis  before  the  A. Ph. A.,  1901, 
Sept.  23d. 
