Am.  Jour.  Pharm. 
May,  1898. 
Hydrogen  Dioxide. 
233 
measuring,  etc.,  as  to  any  other  cause  and  it  would  seem  that,  under 
such  conditions  as  are  at  all  likely  to  occur  in  practice,  the  method 
can  be  depended  upon  to  give  very  serviceable  results. 
Estimation  in  presence  of  larger  quantities  of  glycerin. — In  view  of 
the  fact  that  hydrogen  dioxide  is  sometimes  used  medicinally  in 
combination  with  glycerin,  and  that  assays  of  such  mixtures,  con- 
taining comparatively  large  proportions  of  glycerin,  may  become 
necessary,  trials  were  made  to  determine  the  comparative  accuracy 
of  the  several  methods  under  such  conditions.  A  10  3  volume  solu- 
tion was  used  and  the  results  are  stated  in  volume  strength. 
Grammes  of 
glycerin  added  to 
1  c.c.  of  solution. 
Permanganate 
method  (titr.) 
Permanganate 
method  (gasom.) 
Hypochlorite 
method. 
Thiosulphate 
method. 
0-5 
io-8 
I' 
11-25  ;  n-2 
9'95  ;  9'8 
10-5;  1055 
10-3  ;  10-3 
2" 
1225 
10-5  ;  10-4 
10-35 
It  is  seen  that  the  thiosulphate  method  is  first  and  the  hypochlo- 
rite second  in  point  of  accuracy,  while  the  two  others  are  worthless. 
With  permanganate  the  reaction  is  at  first  extremely  slow,  but  it 
gradually  quickens  and  the  permanganate  is  decolorized  as  rapidly 
as  it  is  added  until  near  the  end,  when  it  again  becomes  very  slow. 
Summary — The  titrimetric  permanganate  method  is  accurate  and 
reliable  for  the  valuation  of  solutions  containing  only  mineral  acids 
and  their  alkali  salts  as  impurities.  With  solutions  containing 
glycerin  or  boroglycerin  in  quantities  below  5  per  cent,  the  results 
are  but  slightly  raised  ;  in  presence  of  larger  amounts  the  method  is 
inapplicable.  Ethereal  solutions  give  results  a  little  too  high. 
Salicylic  acid  interferes  seriously,  even  in  small  quantity.  The  use 
of  this  method  should  be  avoided  whenever  organic  matters  are  sus- 
pected to  be  present.  Applied  gasometrically,  the  permanganate 
method  is  unreliable  in  all  cases. 
The  hypochlorite  method  gives  serviceable  results  in  absence  of 
preservatives.  In  their  presence  they  are  liable  to  fall  too  low,  and 
with  ethereal  solutions  they  are  too  high. 
The  thiosulphate  method  is  simple,  rapid  and  accurate,  and  its 
accuracy  is  not  lessened  by  the  presence  of  the  usual  preservative 
agents,  nor  by  large  quantities  of  glycerin.  It  is  applicable  in  all 
cases,  so  far  as  known.   It  can  be  said  of  gasometric  determinations 
