294 
EXAMINATION  OF  IODIDE  OF  POTASSIUM. 
Solution  of  Coit.  Sub. 
Percentage  of 
left  after  the  exami- 
Pure 
nation — grains. 
lod.  of  Potass. 
Impurities. 
0 
100 
0 
45 
95.5 
4.5 
52 
94.2 
5-2 
248 
75.2 
24.8 
144 
85.6 
14.4 
155 
84.5 
15.5 
Iodide  of  Potassium. 
1.  From  Atkinson  &  Biggar,  London 
2.  Sold  as  an  English  article 
3.  do.       Large  crystals 
4.  Same  bottle,  smallest  crystals 
5.  French  from  Conrad,  Paris 
6.  The  same 
In  order  to  find  the  amount  of  chloride  of  potassium,  the  pre- 
sence of  which  was  shown  by  tests,  of  the  specimens  No.  3  and 
4,  100  grs.  were  precipitated  by  nitrate  of  silver,  the  precipitate 
treated  with  ammonia  and  the  filtered  liquor  neutralized  with 
nitric  acid,  to  separate  the  chloride  of  silver.  Thus  it  was  ob- 
tained of 
No.  3.  134  grs.  Agl,  and  3  grs.  Ag  CI,  corresponding  with  94.72  per  ct.  K  I,  1.57  KCI. 
4.     99  "  16  "  69.75         "  8.32 
The  chloride  found  is  but  a  fraction  of  the  amount  of  impuri- 
ties, therefore  some  other  body  must  be  present.  I  have  satis- 
fied myself  about  the  absence  of  carbonates,  but  could  not  pro- 
ceed any  further  as  the  article  was  returned  to  the  vender.  It  is 
proper  to  state  still  further  that  the  small  crystals  were  very 
damp  ;  some  of  them  lost  at  80°,  2  per  ct. ;  and  at  212°,  2  more 
per  centage  of  moisture,  some  of  the  large  crystals  at  212°  not 
quite  1  per  cent. 
£0  grs.  of  specimen  Nos.  5  and  6  were  precipitated  by  sulphate  of 
copper  under  addition  of  sulphurous  acid.  The  obtained  proto- 
iodide  of  copper  weighed  19  grs.  equal  to  16.55  grs.  or  82.75  per 
ct.  of  pure  iodide  of  potassium.  The  fine  unsuspicious  appear- 
ance of  this  chemical  induced  me  to  prove  the  correctness  of  the 
above  examinations  by  ascertaining  directly  the  quantity  of 
iodine.   I  have  not  searched  yet  for  the  nature  of  the  impurities. 
From  the  foregoing  statements  it  appears  that  according  to 
Marozeau's  treatment  invariably  a  larger  per  centage  of  pure 
iodide  of  potassium  is  found,  being,  however,  not  over  1  or  1.5 
per  cent,  if  the  impurities  are  equally  diffused  among  the  pure 
iodide,  and  some  practice  enables  the  operator  to  find  the  right 
point  where  to  stop  adding  the  solution  of  corrosive  sublimate. 
The  greatest  difference  among  the  above  is  with  No.  4,  being 
about  5.6  per  cent,  which,  however,  seems  to  me  can  be  explained 
by  the  crystals  not  being  an  even  mixture  of  the  iodide  with  the 
impurities,  as  can  be  judged  from  the  analysis  No.  3  of  the 
