Am,/a0nuyi£9*rm"}     Chemical  Observations  on  Tartar  Emetic.  29 
of  the  salt  and  extraction  of  the  mass  with  water  failed  for  the  same 
reason.    The  plan  which  answered  best  was  to  precipitate  a  solution  of 
the  salt  with  ammonia,  which  effected  the  removal  of  nearly  the  whole 
of  the  antimony  as  antimonious  hydrate,  a  small  quantity  being  sub- 
sequently separated  during  the  evaporation  of  the  ammoniacal  nitrate. 
Since  ammonium  and  potassium  tartrates  remain  dissolved  it  is  nec- 
essary before  the  potassium  can  be  estimated  to  ignite  the  residue  left 
after  evaporation.    This  is  accomplished  at  a  low  red  heat,  and  the 
carbonaceous  mass  is  boiled  and  washed  with  water,  and  the  potassium 
finally  precipitated  as  platinochloride  in  the  usual  manner,  or  more 
simply,  the  solution  may  be  titrated  with  an  acid  solution  of  known 
strength.    Working  in  this  way,  1*7408  gram  of  tartar  emetic  yielded 
an  ash  containing  0*3615  gram  of  potassium  carbonate;  the  quantity 
of  potasbium  carbonate  calculated  from  the  formula  SbOKC4U406, 
JH20  is  0*3617  gram.    Having  thus  established  the  correspondence  of 
the  crystals  with  the  formula  (SbOKC4H406,JH20)  they  were  used  in 
the  subsequent  experiments.    Trials  were  first  made  with  solutions  of 
tartar  emetic  acidified  with  hydrochloric  acid,  which  were  titrated  with 
a  decinormal  solution  of  iodine,  using  starch  as  an  indicator.  The 
quantity  of  hydrochloric  acid  added  was,  in  each  experiment,  exactly 
sufficient  to  form  potassium  chloride  from  the  tartar  emetic.  The 
dilution  of  the  solution  was  varied,  but  without  appreciably  affecting 
the  results.    Although  some  of  the  experimental  data  recorded  below 
agree  fairly  well  with  the  calculated  numbers,  yet  the  method  cannot 
be  trusted,  on  account  of  the  absence  of  any  sharp  termination  of  the 
reaction,  which  lags  increasingly  with  each  addition  of  iodine.  This 
difficulty  sufficiently  explains  the  discrepancies  in  the  results,  which 
are  here  recorded  : — 
Tartar  emetic  Dilution  of         Tartar  emetic 
taken.                                        the  liquid.  found. 
0  10                                                 40  0*098 
0*10                                                 60  0*097 
0  10                                                 80  0  099 
020                                                160  0  1 95 
0-20                                                100  0  195 
Experiments  were  next  made  with  solutions  of  tartar  emetic  to 
which  sodium  bicarbonate  had  recently  been  added.  When  this  salt 
is  dissolved  in  a  solution  of  tartar  emetic  no  visible  change  is  observed 
at  first,  but  after  the  lapse  of  a  few  minutes  the  liquid  becomes  turbid, 
