296       PREPARATION  OF  CITRATE  OP  IRON  AND  QUINIA. 
up  with  water  and  again  stirred.  When  the  precipitate  shall 
have  subsided,  decant  the  supernatant  liquid  and  again  fill  up  the 
vessel  with  water  and  stir  the  mixture.  Repeat  this  washing  the 
precipitate  four  times,  or  until  the  washings  give  no  cloudiness 
with  solution  of  chloride  of  barium.  Then  transfer  the  oxide  to 
a  strainer,  and  drain  it  till  it  can  be  heaped  up  with  a  spatula. 
It  will  then  weigh  about  34j,  and  if  dried  by  sulphuric  acid  under 
a  bell  glass  will  yield  174  grains  of  terhydrated  sesquioxide  of 
iron.  Or  if  dried  on  a  sand  bath  and  subjected  for  15  minutes  to 
a  red  heat,  it  will  yield  134  grains  of  sesquioxide  of  iron. 
The  yield,  by  calculation,  from  480  grains,  Fe  0,  S03+ 7  HO, 
should  be  184.747  grs.  Fe2  03+3  HO,  or  138.128  grs.°  Fe2  03. 
It  is  better  to  heat  the  sulphuric  acid  and  water  first,  and  to 
add  the  sulphate  of  iron  immediately  after  the  nitric  acid,  as  this 
avoids  loss  of  nitric  acid  by  evaporation.  It  is  better  also  to  add 
the  sulphate  at  once,  and  in  lumps  as  it  occurs,  as  the  effervescence 
is  then  moderate,  and  there  is  less  risk  of  loss  by  spilling,  or  the 
spitting  of  the  effervescence.  Hence,  also,  the  utility  of  a  pro- 
portionately deep  vessel,  and  a  quantity  of  water  so  small  as  not 
to  cover  or  embrace  the  whole  of  the  salt  at  once.  The  peroxi- 
dation is  probably  accomplished  as  rapidly  as  the  salt  can  be 
dissolved,  for  I  have  never  been  able  to  get  a  clear  indication  of 
protoxide  in  the  solution,  even  when  tested  before  all  effervescence 
had  ceased.  During  the  prolonged  heating,  however,  nitrous 
fumes  are  given  off  even  for  many  hours,  and  long  after  the  solu- 
tion of  the  sulphate  of  peroxide  of  iron  is  perfect.  An  hour  or 
an  hour  and  a  half  of  heating  gets  rid  of  most  of  this  entangled 
or  dissolved  nitric  oxide,  and  in  practice  seems  quite  sufficient  to 
prevent  its  interference  in  the  precipitation.  The  large  dilution 
of  the  solution  before  precipitation  is  useful,  for  it  renders  the 
resulting  precipitate  more  readily  soluble,  It  is  better  to  wash 
the  oxide  by  decantation,  because  it  is  better,  more  easily,  and 
much  more  speedily  done.  By  decantation,  with  a  piece  of  India 
rubber  tubing  as  a  syphon,  the  washing  can  be  well  performed  in 
two  hours,  without  any  contact  of  air  till  the  magma  comes  to  be 
drained  on  the  strainer.  Here  it  invariably  gets  a  little  carbonic 
acid. 
In  making  the  oxide  hastily  as  an  antidote,  the  prolonged 
heating,  the  large  dilution  of  the  solution,  and  three  of  the  wash- 
