CRITICAL  NOTE  ON  PERMANGANATE  OF  POTASSA.  395 
tate,  and  the  supernatant  liquid  has  a  reddish  color,  gradually 
passing  to  brownish,  and,  if  allowed  to  stand,  becoming  perfectly 
clear. 
On  acidifying  the  original  solution  with  nitric  acid  no  change 
is  produced,  except  at  a  high  temperature  when  bromine  is 
liberated.  If,  however,  a  drop  or  two  of  the  saturated  solution 
of  potass,  permang.  is  added  to  the  acid  solution,  or  the  liquid 
acidified  after  the  addition  of  potass,  permang.,  bromine  is  imme- 
diately liberated  without  the  application  of  heat.  At  the  same 
time  similar  reactions  are  obtained  as  when  the  potass,  permang. 
is  added  to  the  neutral  solution. 
In  the  dilute  neutral  solution  potass,  permang.  produces  no 
effect ;  but  on  the  addition  of  a  drop  or  two  of  nitric  acid  the 
liquid  assumes  a  brownish  color,  due  to  the  liberation  of  free 
bromine. 
The  rationale  of  these  reactions  is  that  the  permanganate 
gives  up  part  of  its  oxygen  to  the  halogens,  converting  them  into 
chlorates,  iodates  or  bromates,  and  brown  hydrated  dentoxide  of 
manganese  is  precipitated.  At  the  same  time  a  small  quantity 
of  chlorine,  iodine  or  bromine  is  liberated,  the  two  latter  in 
sufficient  quantity  to  give  color  to  the  supernatant  liquid  even  in 
very  dilute  solutions.  The  reactions  are  of  course  much  more 
energetic  in  strong  solutions  than  in  dilute  ones.  In  fact,  in  the 
case  of  dilute  solution  of  chloride  of  sodium  it  is  so  slight  as  not 
to  be  discernible. 
From  the  foregoing  experiments  it  will  be  seen  that  with 
moderately  strorig  ^q\vX\qy^^  of  the  three  halogens  under  considera- 
tion, the  reactions  with  permang.  potass,  are  so  nearly  alike  in 
all  respects  that  it  would  be  unsafe  to  place  any  confidence  in 
the  indications. 
The  method  therefore  is  only  applicable  to  very  dilute  solu- 
tions, and  even  then  care  must  be  taken  that  the  permang. 
potass,  is  also  in  very  dilute  solution,  and  added  droip  hy  drop,  as 
any  excess  beyond  what  the  iodide  or  bromide  would  decompose 
would  be  apt  to  give  the  liquid  its  own  characteristic  tint. 
Moreover,  the  method  is  applicable  only  to  simple  solutions  of 
haloid  salts,  and  its  use  implies  some  previous  knowledge  by  the 
operator,  both  of  the  composition  and  the  strength  of  the  solu- 
