82  DETERMINATION  OF  GRAPE-SUGAR,  ETC. 
its  color  very  slowly  in  the  cold,  more  rapidly  between  122° 
and  140°  F.,  but  very  rapidly  between  140Q  and  176°  F. 
If  a  few  drops  of  the  solution  of  this  reagent  be  poured  into 
a  solution  of  grape-sugar,  heated  to  140°  F.,  and  then  well 
shaken,  the  yellow  color  first  produced  very  soon  disappears  (at 
176°  F.  almost  instantaneously).  When  the  coloration  is  re- 
produced by  the  addition  of  a  further  portion  of  the  reagent,  it 
always  disappears  again  as  long  as  grape-sugar  is  present.  To- 
wards the  end,  the  decolorization  takes  place  more  slowly,  and 
is  then  facilitated  by  heating  the  fluid  to  176°  F.  If  the  fluid 
retains  its  color  at  this  temperature,  all  the  grape-sugar  is  de- 
stroyed. 
The  reagent  is  remarkably  sensitive;  the  yellow  color  which 
persists  at  last  in  consequence  of  the  addition  of  an  excess  of 
the  reagent,  may  be  removed  by  a  few  drops  of  grape-sugar. 
3.  Dextrine,  prepared  by  roasting  starch,  has  no  action  upon 
this  reagent,  even  when  it  has  been  treated  with  muriatic  acid 
under  the  same  circumstances  in  which  cane-sugar  is  converted 
into  grape-sugar. 
4.  If  a  solution  of  cane-sugar  in  40  times  its  weight  of  water, 
to  which  55  per  cent,  of  the  weight  of  the  sugar  of  concentra- 
ted muriatic  acid  has  been  added,  be  heated  in  the  water-bath 
to  129° — 131°  F.,  all  the  cane-sugar  is  converted  into  grape- 
sugar.  If  this  solution  be  then  neutralized  with  carbonate  of 
soda  (an  excess  of  which  has  no  influence),  it  behaves  like  a  so- 
lution of  pure  grape-sugar. 
In  order  to  apply  these  facts  to  the  determination  of  grape- 
sugar,  cane-sugar,  and  dextrine  in  their  mixtures,  it  was  first 
to  be  ascertained  what  quantity  of  ferridcyanide  of  potassium 
is  necessary  to  decompose  a  certain  weight  of  grape-sugar. 
Three  experiments  with  a  normal  solution  of  ferridcyanide  of 
potassium,  proved  that  on  an  average  it  required  10,980  mil- 
ligrms  of  the  reagent  for  the  decomposition  of  1000  milligrms. 
of  cane-sugar  which  had  been  converted  into  grape-sugar  by 
means  of  muriatic  acid,  or  10,980  grms.  of  the  salt  to  1  grm. 
of  cane-sugar. 
A  test-fluid  was  then  prepared  containing  in  100  cubic  cen- 
tims.  10,980  grms.  of  ferridcyanide  of  postassium  and  5|  grms. 
