Amjune,r'i9Pi4arm'}   Volumetric  Estimation  of  Sulphates.  25$ 
equivalent  to  0.4066  gm.  BaCL  +  2H20.  Second  lot  of  BaCl2 : 
10  c.c.  yielded  0.3886  gm.  BaS04,  equivalent  to  0.4067  gm.  BaCL 
+  2H20.  Theory  demanded  0.4072  gm.  BaCL,  +  2H2Q.  Error 
=  0.1  per  cent. 
Ordinarily  about  0.16  gm.  S04  should  be  taken.  If,  however, 
the  proportion  of  SC4  is  so  small  that  0.16  gm.  cannot  be  taken 
without  increasing  unduly  the  concentration  of  the  other  ions,  it  is 
necessary  to  take  into  account  the  error  arising  from  the  solubility 
of  the  BaCr04,  which,  under  the  conditions,  is  represented  by  0.3  c.c. 
n/io  Na2S203  per  40  c.c.  of  the  filtrate.  For  example,  10  gms. 
table  salt  were  dissolved  in  sufficient  water  to  measure  100  c.c. 
Volumetric  and  gravimetric  estimations  were  made  on  10  c.c. 
portions.  40  c.c.  of  the  filtrate  required  1.1  c.c.  n/io  Na2S20;r  The 
correction  having  been  deducted,  0.0064  gm.  S04  was  indicated. 
BaS04  found:  0.0152  gm.,  equivalent  to  0.0062  gm.  S04. 
Table  I  presents  a  large  part  of  the  experimental  data  and  is  self- 
explanatory. 
Notes  Supplementary  to  Table  I. 
a.  Carefully  recrystallized  and  dehydrated  by  gentle  ignition;  purity  also 
established  gravimetrically  in  the  course  of  some  previous  work. 
.  b.  Cu-  does  not  interfere,  provided  care  is  taken  to  prevent  the  premature 
formation  of  cuprous  salt.  Should  this  occur,  the  titrating  fluid  becomes 
cloudy.  Shaking  and  standing  may  clear  it.  The  n/io  Na^S-Oa  should  be 
added  in  small  portions  with  vigorous  shaking.  The  first  disappearance  of 
the  blue  color  is  taken  as  the  end.  The  titration  may  also  be  carried  out 
without  starch  indicator,  the  thiosulphate  being  run  in  until  there  is  no  further 
change  in  color  and  the  liquid  acquires  a  slight  permanent  opalescence. 
c.  Mg  occasions  a  positive  error  variable  between  1  and  2  per  cent.  The 
substitution  of  caustic  soda  for  ammonia  does  not  reduce  this  error. 
d.  Two  independent  experiments. 
e.  Although  SrCrCXt  is  readily  soluble  under  the  conditions  of  the  pro- 
cedure, the  slight  solubility  of  SrSOi  greatly  limits  the  application  of  the 
method  in  the  presence  of  Sr. 
/.  Less  the  correction  0.3  =  2  c.c. 
g.  The  salt  was  badly  effloresced. 
h.  The  study  of  this  element  led  to  the  addition  of  the  ammonia  and 
acetic  acid  mixed.  Alumina  once  precipitated  by  ammonia  redissolves  with 
great  difficulty  in  so  weak  an  acid  as  acetic.  Further,  precipitated  alumina 
occludes  soluble  chromate.  Under  the  conditions  of  the  procedure  the 
aluminium  remains  in  solution  or,  at  most,  the  liquid  becomes  slightly 
opalescent. 
i.  The  usual  formula  with  7H2O  was  assumed;  hence  the  weight  taken. 
The  salt  actually  contained  6H2O. 
