228  Hydrogen  Dioxide. 
the  permanganate  method.  A  saturated  solution  of  sodium  chloride 
was  used  in  all  cases  for  filling  the  gasometer.  Two  c.c.  of  the 
sample  are  admitted  into  the  measuring  tube,  the  funnel  tube  rinsed 
with  a  little  water,  this  also  let  in  and  then  20  c.c.  of  the  chlorinated 
lime  solution  added.  From  this  point  the  procedure  is  the  same  as 
explained  under  the  permanganate  method. 
With  solutions  containing  in  addition  to  hydrogen  dioxide  only 
small  quantities  of  mineral  acids  and  alkali  salts,  quite  satisfactory 
results  were  obtained  by  this  method,  but  in  the  cases  of  solutions 
to  which  any  of  the  various  preservatives  had  been  added,  there 
was  a  decided  tendency,  almost  throughout,  toward  too  low  results. 
No  appreciable  error  seems  to  be  introduced  through  the  solubility 
of  oxygen  in  the  liquid,  above  which  the  gas  is  collected,  as  no 
higher  results  were  obtained  when  the  liquid  had  previously  been 
saturated  with  oxygen. 
Titrimetric  Thiosulphate  Method.- — This  was  first  recommended  by 
C.  T.  Kingzett  {Jour.  Client.  Soc,  1880,  Vol.  37,  p.  802),  and  later 
by  H.  Thorns  {Arch.  Pharm.,  1887,  Vol.  35,  pp.  335-343).  It  is 
based  on  the  fact  that  hydrogen  dioxide  liberates  a  definite  quantity 
of  iodine  from  iodides  in  acid  solution,  the  iodine  being  titrated 
with  sodium  thiosulphate,  in  the  usual  manner. 
H202  -f  2KI  -I-  H2S04  =  I2  K2  +  S04  +  2H20 ; 
I2  -f  2Na2S203  =  2NaI  +  Na2S4Oe. 
Thorns  proceeds  by  warming  0-5  gramme  of  potassium  iodide,  1 
gramme  of  20  per  cent,  sulphuric  acid  and  50  c.c.  of  water  to  400  C, 
adding  0-5  c.c.  of  the  hydrogen  dioxide  solution,  shaking  about  five 
N 
minutes,  and  titrating  the  liberated  iodine  with  —  thiosulphate, 
10 
then  warming  again  to  400,  again  titrating  the  appearing  iodine, 
and  repeating  until  no  further  liberation  of  iodine  takes  place. 
Carried  out  in  this  way  the  method  is  rather  slow  and  tedious, 
requiring  at  least  fifteen  minutes  for  completion ;  but,  since  it 
was  stated  to  be  accurate,  the  writer  decided  to  include  it  in 
his  experiments,  and  found  later,  that  by  omitting  the  greater 
part  of  the  water,  so  as  to  reduce  the  bulk  of  the  solution  to  about 
15  c.c.  for  the  above-mentioned  quantities,  heating  might  be  dis- 
pensed with,  and  the  reaction  would  nevertheless  be  ended  within 
five  minutes.    The  final  result  was  found  to  be  the  same,  whether 
