AmDIe°cUIri8?i!arm-}  Ferric  Phosphate  and  Pyrophosphate.  589 
(2)  .  If  the  salt  to  be  tested  purports  to  be  pyrophosphate  of  iron 
U.  S.  P.  a  portion  of  the  alkaline  filtrate  "  A  "  is  tested  first  for 
contaminating  phosphates,  by  the  method  described  above,  only 
using  instead  of  an  excess  of  solution  of  magnesium  sulphate,  two 
to  four  drops  of  the  same.  In  this  way  magnesium  pyrophosphate, 
which  is  precipitated  at  the  first  admixture  of  the  reagent,  redis- 
solves  in  the  excess  of  alkali  pyrophosphate,  while  phosphates,  if 
present,  are  precipitated  as  magnesium  ammonium  phosphate.  A 
good  preparation  should  not  yield  more  than  a  very  slight  turbidity 
of  phosphate.  If  a  precipitate  is  formed,  it  is  filtered  after  five  min- 
utes, washed  and  converted  into  the  silver  salt.  In  this  way  the 
admixture  of  one  part  of  phosphate  to  nine  parts  of  pyrophosphate 
of  iron  could  be  easily  detected. 
(3)  .  Another  part  (3  cc.)  of  the  alkaline  filtrate  "A"  is  now 
tested  for  pyrophosphates  by  neutralizing  with  acid  and  then 
adding  an  excess  of  five  to  six  drops  of  acetic  acid  (36  per  cent.) 
and  1-5  cc.  magnesium  sulphate  solution.  On  boiling  the  clear 
solution,  magnesium  pyrophosphate  is  precipitated,  if  present,  while 
if  the  salt  was  a  phosphate  no  precipitation  occurs.  For  further 
identification  the  precipitate  can  be  filtered  hot  and  washed  with 
hot  water  till  the  washings  are  neutral.  If  it  be  dissolved  in  a  drop 
or  two  of  nitric  acid,  reprecipitated  with  ammonia  and  redissolved 
with  a  drop  of  acetic  acid,  silver  nitrate  will  produce  white  earthy 
pyrophosphate  of  silver.  A  very  slight  yellowish  tinge  may  occur 
from  insufficient  washing,  but  there  is  no  possible  mistaking  of  the 
precipitate,  the  citric  acid  having  been  removed  from  the  solution. 
For  further  particulars  of  this  test  reference  may  be  made  to  the 
next  article.  In  this  manner  the  presence  of  pyrophosphate  is 
proved  beyond  doubt,  the  demonstration  resting  on  positive  reac- 
tions. The  whole  examination  can  be  completed  very  nearly  as 
quickly  as  the  present  U.  S.  P.  test.  Samples  have  been  tested 
showing  contaminations  with  phosphates  which  we  have  had  no 
hesitation  in  accepting,  having  proof  positive  of  the  pyrophosphate 
present.  The  samples  have  also  been  examined  which  were  sold 
as  pyrophosphate  U.  S.  P.,  but  which  proved  to  be  phosphate,  by 
the  U.  S.  P.  test,  the  white  precipitate  of  silver  citrate  seemed  to  be 
proof  of  pyrophosphate.  The  yellowish  and  yellow  tints  also  ob- 
tained by  fractional  precipitation  were  no  unusual  accompaniments 
of  the  pyrophosphate  of  silver ;  but  by  the  test  described  no  pre- 
