148  Separation  of  Oxymethylanthraquinones.  {AmApSr'i9i5arm' 
in  a  separatory  funnel,  first  with  25  Cc.  portions  of  5  per  cent, 
ammonium  carbonate  until  the  washings  were  colorless  or  but  faintly 
colored ;  then  with  similar  portions  of  5  per  cent,  sodium  carbonate 
until  the  washings  became  pink  colored ;  and,  finally,  with  5  per  cent, 
sodium  hydroxide,  thus  removing  unidentified  anthraquinone  deriva- 
tives, emodin,  and  chrysophanic  acid  in  the  order  named.  The  three 
aqueous  solutions  were  then  separately  acidified  and  shaken  out 
with  ether,  and  the  ether  residues  examined  directly  or  after  re- 
crystallization  from  alcohol. 
The  differentiation  of  the  several  residues  was  made  chiefly  on 
the  basis  of  their  melting-points  and  color  reactions.  After  testing 
the  residues  with  various  reagents  for  distinctive  color  reactions,  it 
was  found  that  the  most  striking  differentiation  was  obtained  by 
treating  with  concentrated  sulphuric  acid,  concentrated  nitric  acid, 
and  water  successively  in  the  following  manner : 
To  the  residue  in  a  white  porcelain  capsule  were  added  4  to  5 
drops  of  concentrated  sulphuric  acid,  then  1  to  2  drops  of  concen- 
trated nitric  acid,  and,  finally,  about  1  Cc.  of  water.9  The  initial 
color  with  sulphuric  acid  and  the  behavior  on  dilution  with  water 
constitute  the  characteristics  of  the  test.  The  residue  from  a  few 
drops  of  an  ethereal  solution  of  the  color  in  question  is  sufficient  for 
the  test. 
Chrysarobin. 
The  procedure  just  described  was  first  applied  to  commercial 
"  chrysophanic  acid."  Since  this  substance  is  largely  chrysarobin, 
about  3  grammes  of  the  powder  were  dissolved  in  100  Cc.  of  dilute 
(10  per  cent.)  sodium  hydroxide  and  oxidized  by  passing  air  through 
the  solution  for  one  hour.  The  solution  was  then  filtered,  the  filtrate 
acidified  with  dilute  sulphuric  acid,  and  a  hot  extraction  made  with 
benzol. 
Ammonium  and  Sodium  Carbonate  Fractions. — A  little  color  was 
removed  by  each  of  these  reagents,  but  the  fractions  were  not  further 
examined. 
Sodium  Hydroxide  Fraction. — The  remaining  color  was  removed 
completely  from  the  benzol  in  one  washing.  The  color  was  passed 
into  ether  and  the  ether  evaporated  at  a  gentle  heat.  The  residue 
was  amorphous ;  golden  yellow  in  color ;  soluble  in  hot,  but  difficultly 
9  These  reagents  are  the  same  as  used  by  Cripps  and  Dymond  in  their 
test  for  aloes. — Ber.  Chem.  Ges.,  18,  200  (1885). 
