IODINE  IN  MINERALS,  PLANTS  AND  ANIMALS. 
443 
oxide  of  barium.  It  is  stated  that  a  millionth  part  of  iodine 
may  be  detected  by  this  test. 
Casaseca  considers  that  the  use  of  pure  nitric  acid  with  starch 
is  quite  as  safe  as  peroxide  of  barium.  In  order  to  detect  iodine 
in  mineral  water  he  extracts  the  dry  saline  residue  with  acetic 
ether,  in  which  the  alkaline  iodides  are  soluble,  while  the  accom- 
panying bromides,  chlorides,  sulphites  and  hyposulphites  are  inso- 
luble. The  solution  is  evaporated,  the  residue  dissolved  in  water, 
and  treated  with  nitric  acid  and  starch  paste. 
E.  Marchand  tests  ashes  and  mineral  water  for  iodine  by  mix- 
ing the  aqueous  solution  in  a  stoppered  vessel  with  dry  starch, 
excess  of  hydrochloric  acid,  and  about  0-0001  grm.  of  chromate 
of  potash.  After  shaking  the  mixture  and  allowing  it  to  rest,  a 
rose  color  is  produced,  when  ten  cubic  centimeters  of  liquid  are 
used,  containing  only       of  iodine. 
Rabourdin  takes  advantage  of  the  fact,  that  free  iodine  dis- 
solves in  chloroform  with  a  violet  color.  If  ten  grm.  of  a  liquid 
containing  of  its  weight  of  iodide  of  potassium  are  mixed  with 
two  drops  of  nitric  acid,  fifteen  or  twenty  drops  of  sulphuric  acid, 
and  one  grm.  of  chloroform,  the  latter  acquires  a  violet  color 
when  the  mixture  is  shaken.  This  color  is  more  intense  in  pro- 
portion to  the  quantity  of  iodine  present,  and  may  serve  for  the 
approximative  estimation  of  iodine  by  comparison  with  similar 
solutions  of  known  value.  Chloroform  abstracts  iodine  even  from 
an  aqueous  solution;  when  it  contains  ether  the  color  produced 
is  not  violet,  but  more  or  less  red  in  proportion  to  the  quantity 
of  ether. 
Lassaigne  gives  the  preference  to  chloride  of  palladium  as  a 
reagent  for  iodine,  which  is  more  delicate  than  starch,  and  which 
has  the  advantage  of  indicating  the  presence  of  iodine  in  saline 
compounds  when  starch  no  longer  giver  a  result.  He  states  that 
he  has  detected  two-millionths  of  iodide  of  potassium  in  2000 
grm.  of  water  by  the  brown  color  which  chloride  of  palladium 
communicates  to  the  liquid.  After  the  lapse  of  twenty-four  or 
thirty-six  hours  flocks  of  iodide  of  palladium  separate,  which, 
when  mixed  with  a  little  moist  silica  and  heated  in  a  tube,  evolve 
violet  iodine  vapor. 
Winckler  recommends  as  the  most  delicate  test  for  iodine,  next 
to  chloride  of  palladium,  nitrite  of  potash  or  soda  ;  he  found  this 
