ON  A  NEW  TEST  FOR  IODIDES. 
143 
suspected  to  contain  an  iodide  is  mixed  with  starch-paste  and 
acidified  with  hydrochloric  acid  ;  a  solution  of  nitrite  of  potash  is 
then  added,  when,  if  much  iodine  be  present,  a  dark  blue  color  will 
be  instantly  produced;  if  a  very  small  quantity  only,  as  for  instance 
the  two  or  three  millionth  part,  then  a  few  seconds  elapse  before 
the  blue  color  makes  its  appearance.  In  this  manner  I  have  de- 
tected the  l-4000500th  part  of  iodine  dissolved  in  water  as 
iodide  of  potassium.  It  will  be  seen  that  the  test  admits  of  a 
degree  of  delicacy  not  attainable  by  any  of  the  other  methods  for 
detecting  iodides,  as  well  as  being  at  the  same  time  free  from  the 
disadvantages  to  which  they  are  more  or  less  subject ;  as.  for  in- 
stance, in  the  employment  of  chlorine,  which,  unless  added  very 
carefully  to  a  liquid  containing  a  trace  of  an  iodide  only,  is  al- 
most sure  to  afford  a  negative  result,  from  the  chlorine  combin- 
ing with  the  iodine,  and  so  preventing  its  acting  on  the  starch. 
The  same  error  may  also  arise  by  the  use  of  nitric  acid  should 
the  suspected  liquid  contain  a  large  amount  of  chlorides. 
I  will  now  detail  the  two  cases  in  which  I  have  applied  this  test; 
in  the  one  for  the  purpose  of  detecting  iodine  in  cod-liver  oil,  the 
object  being  to  see  how  small  a  quantity  of  the  oil  would  suffice  ; 
in  the  other,  for  the  purpose  of  detecting  iodine  in  marine  vege- 
tation. One  ounce  of  ordinary  brown  cod-liver  oil  was  saponified 
by  a  concentrated  solution  of  caustic  potash,  and  then  carbonized 
in  an  iron  spoon  over  an  open  fire ;  the  residue  was  removed  into 
a  covered  porcelain  crucible,  and  strongly  heated,  so  as  effectu- 
ally to  destroy  all  organic  matter,  and  when  cold  was  digested 
with  a  small  quantity  of  water,  and  thrown  upon  a  filter  ;  the  fil- 
trate, being  acidified  with  hydrochloric  acid,  was  then  mixed  with 
starch-paste,  and  tested  with  nitrite  of  potash,  which  almost  im- 
mediately produced  a  pale  plum  color. 
Sea-water  contains  so  small  an  amount  of  iodine,  that  it  is  ex- 
ceedingly difficult  to  detect  even  a  trace  of  it  in  the  mother-li- 
quor from  several  pounds  of  water.  Minute  as  this  quantity 
must  be,  it  is  nevertheless  collected  and  assimilated  by  many  ma- 
rine plants,  and  the  following  experiments  enable  us  to  demon- 
strate its  presence  in  their  juices.  If  wre  take  a  thin  transverse 
sectional  slice  of  the  stem  of  the  Fucus  laminaria  digitata, 
moisten  it  with  a  little  starch-paste  and  dilute  hydrochloric  acid, 
and  examine  it  by  the  aid  of  the  microscope,  we  shall,  upon  add- 
ing a  drop  of  a  solution  of  nitrite  of  potash  to  the  same,  be  able 
