AmNovliri9Pi8arm'  ^       Determination  of  Morphine.  789 
besides  leaving  a  complete  uncertainty  as  to  whether  any  morphine 
present  will  be  in  the  solid  or  in  the  liquid  portion.  Barium  hy- 
droxide was  found  very  effective  in  rendering  these  extracts  amena- 
ble to  filtration,  and  it  has  the  added  virtue  of  completely  dissolving 
any  morphine,  no  matter  in  what  form  it  may  be  present.  The  sub- 
sequent separation  of  morphine  in  a  form  sufficiently  pure  for  titra- 
tion was  effected  by  an  adaptation  of  J.  B.  Williams's  method.  The 
errors  shown  to  accompany  the  precipitation  of  morphine  by  ammo- 
nium hydroxide  in  the  presence  of  alcohol  (see  Part  I)  render  such 
a  mode  of  procedure  most  undesirable. 
Detailed  Description  of  Method. 
The  material  to  be  examined,  if  a  solid,  should  be  divided  as 
finely  as  possible.  Weigh  a  sample  of  6  Gm.  and  transfer  it  to  a 
100  Cc.  graduated  flask.  Add  approximately  2  Gm.  of  pure  cal- 
cium carbonate  and  mix  by  gently  shaking,  then  add  20  Cc.  of  water 
and  warm  on  the  water  bath  till  a  uniform  thin  paste  has  been 
formed.  While  the  warming  is  in  progress  the  flask  should  be 
loosely  closed  by  a  rubber  stopper.  Cool  the  mixture,  then  add  60 
Cc.  of  a  cold  saturated  solution  of  barium  hydroxide.  Mix  the  con- 
tents of  the  flask  thoroughly  and  allow  to  stand  for  20  to  30  min- 
utes. Dilute  with  water  to  100  Cc,  mix  well,  and  pour  on  a  dry 
15  Cm.  filter  paper. 
Treatment  of  Precipitate  - — Transfer  the  residue  on  the  filter  to 
a  beaker,  add  a  slight  excess  of  hydrochloric  acid,  warm  almost  to 
boiling,  and  filter.  Collect  as  much  as  possible  of  this  filtrate  and 
concentrate  to  about  15  Cc.  Test  this  with  ferric  chloride  for  me- 
conic  acid.  Its  presence  indicates  that  any  morphine  in  the  pills 
being  examined  is  in  the  form  of  opium,  in  which  case  the  method 
of  determination  about  'to  be  described  cannot  be  used,  and  the 
modification  which  will  be  described  in  Part  III  must  be  employed. 
If  no  meconic  acid  is  found  the  course  to  be  followed  is  as  now 
described. 
Treatment  of  Filtrate. — Transfer  50  Cc.  of  the  filtrate  to  a  flask 
graduated  to  contain  55  Cc,  such  as  is  commonly  used  in  the  anal- 
ysis of  syrups  and  sugars.  Add  sulphuric  acid  (diluted  1:5)  very 
cautiously,  till  all  barium  has  been  precipitated  and  the  solution  is 
just  acid  to  litmus.  Dilute  the  contents  of  the  flask  to  the  55  Cc 
mark  and  mix  thoroughly.    Allow  the  precipitate  to  settle  ;  filter 
