AmDec.!:i89ifrm"}  Pyrophosphoric  and  Phosphoric  Acid.  591 
pure  pyrophosphoric  acid  is  slowly  converted  even  in  the  cold  by 
the  nitric  acid  employed  in  the  test. 
The  solubility  of  magnesium  pyrophosphate  in  alkali  pyrophos- 
phate gives  a  convenient  test  for  demonstrating  pyrophosphate  in 
the  presence  of  only  a  small  proportion  (Ca.  10  per  cent.)  of  phosphate, 
even  in  the  presence  also  of  an  acid  like  citric  acid  which  has  a  white 
silver  salt.  If  to  an  ammoniacal  solution  of  the  acids  named,  some 
ammonium  chloride  and  only  a  few  drops  of  magnesium  sulphate 
solution  are  cautiously  added  as  long  as  the  original  precipitate  of 
magnesium  pyrophosphate  redissolves,  magnesium  ammonium 
phosphate  will  slowly  separate  out,  especially  on  rubbing  with  a 
glass  rod.  On  filtering  after  a  little  while  and  boiling  the  filtrate,  a 
heavy  precipitate  would  indicate  pyrophosphate,  which  can  now  be 
tested  with  silver  nitrate  without  the  disturbing  influence  of  either  of 
the  other  acids.  This  process  is  naturally  very  limited,  presupposing, 
as  was  said,  only  a  very  small  percentage  of  phosphates.  If  con- 
siderable phosphate  is  present,  magnesium  pyrophosphate  would  be 
precipitated  together  with  the  phosphate  by  the  addition  of  suffi- 
cient magnesium  sulphate,  the  remaining  alkali  phosphate  being 
insufficient  to  keep  the  magnesium  salt  in  solution.  This  method 
is  best  combined  with  the  previous  method  for  testing  and  identi- 
fying pyrophosphates. 
The  fact  that  magnesium  pyrophosphate  dissolves  in  excess  of 
magnesium  sulphate  and  is  not  precipitated  from  such  a  solution  by 
ammonium  salts,  forms  a  foundation  for  another  well-known  method 
for  separating  the  two  acids.  The  method  was  found  practical, 
but  a  very  large  excess  of  magnesium  sulphate  was  required  for 
dissolving  the  pyrophosphate  of  magnesium.  On  adding  concen- 
trated magnesium  sulphate  solution  (20  G  salt  in  60  cc.  H20)  to  a 
solution  of  sodium  pyrophosphate,  one  volume  of  the  former  solu- 
tion effecting  total  precipitation  of  the  pyrophosphoric  acid,  15 
volumes  were  required  to  redissolve  the  precipitate.  On  the  other 
hand,  by  pouring  the  solution  of  sodium  pyrophosphate  into  con- 
centrated magnesium  sulphate  solution,  an  excess  of  eight  volumes 
of  the  latter  was  required  to  give  a  clear  solution.  This  very  large 
excess  of  concentrated  solution  of  magnesium  sulphate  makes  the 
test  rather  inconvenient ;  but  it  is  reliable  and  otherwise  practical. 
The  quickest  and  most  convenient  method  for  separating  the 
acids  for  qualitative  work  appears  to  me  to  rest  in  their  behavior 
