VOLUMETRIC  ANALYSIS  OF  SUPERPHOSPHATE  OF  LIME.  345 
ON  A  VOLUMETRIC  ANALYSIS  OF  SUPERPHOSPHATE  OF 
LIME. 
By  George  Jones,  F.C.S. 
It  has  for  some  time  been  a  desideratum  of  analytical  agricul- 
tural chemists,  and  especially  of  those  connected  with  the  manu- 
facture of  artificial  manures,  to  be  enabled  by  a  more  speedy 
method  than  the  one  generally  pursued  to  arrive  at  a  correct 
estimation  of  the  amount  of  phosphate  of  lime  existing  in  a  solu- 
ble state  in  the  so-called  "superphosphate."  I  have  therefore 
been  led  to  attempt  a  series  of  experiments  upon  a  process  of 
volumetric  estimation,  by  the  use  of  a  standard  solution  of  am- 
monia;  but  at  the  onset  I  experienced  some  difficulty  in  arriving 
at  any  satisfactory  result  in  consequence  of  the  existence  of  the 
free  or  uncombined  sulphuric  acid,  which  is  invariably  present  in 
ordinary  commercial  superphosphate,  it  being  impossible  to  pre- 
sent an  alkali  without  precipitating  the  phosphate  of  lime. 
The  process  I  therefore  adopt  is  the  following  :  Having  taken 
100  grains  of  the  sample  for  analysis,  it  is  first  of  all  well  mixed 
with  about  50  grains  of  finely  powdered  litharge,  and  introduced 
along  with  a  small  quantity  of  distilled  water  into  a  flask,  and 
boiled  for  about  fifteen  minutes.  The  whole  is  then  made  up  with 
distilled  water  to  7000  grains  (one  deci-gallon),  agitated  well,  and 
thrown  upon  a  filter.  I  then  take  of  the  filtrate,  by  the  use  of  a 
pipette,  1400  grains  (200  septems),  equal  to  20  grains  of  the 
sample,  and  add  thereto  a  little  chloride  of  calcium  solution,  and 
200  septems  of  the  standard  solution  of  ammonia  ;  it  is  then 
made  up  with  distilled  water  to  a  known  bulk — say  2000  grains, 
agitated  well,  and  filtered. 
A  1000  grains  pipette  of  the  filtrate  will  therefore  represent 
exactly  10  grains  of  the  sample,  and  in  this  I  now  proceed  to 
test  for  ammonia,  added  over  and  above  that  required  to  sepa- 
rate the  phosphate  of  lime  in  the  sample.  Two  equivalents  of 
ammonia — being  required  for  every  equivalent  of  the  tribasic 
phosphate  of  lime  precipitated.    Thus — 
CaOfPO6>+2Ca01+2NH4O=(CaO)5PO5  and  2NH4C1. 
The  standard  solution  of  ammonia  I  am  in  the  habit  of  using 
contains  in  every  septum  *01  of  real  ammonia,  and  I  employ  also 
