146  Separation  of  Oxymethylanthraquinones.  {  Am"Ap?SSr'1^5 
anthraquinones.4  This  test  depends  upon  the  production  of  a  red 
color  in  the  aqueous  liquid  when  an  ethereal  solution  of  oxymethyl- 
anthraquinones is  washed  with  dilute  alkali.  This  cannot  be  re- 
garded as  a  specific  test,  since  other  colors,  both  natural  and  synthetic, 
behave  similarly.  The  writer's  experience  with  such  conditions  led 
to  the  outline  of  a  procedure 5  to  distinguish  chrysophanic  acid 
(dioxymethylanthraquinone)  in  the  presence  of  phenolphthalein, 
naematoxylin,  and  curcumin.6  After  hematoxylin  and  curcumin 
have  been  eliminated,  chrysophanic  acid  is  distinguished  from  phenol- 
phthalein by  reducing  both  substances  with  zinc  dust  and  dilute 
sodium  hydroxide,  whereupon  chrysophanic  acid  is  reduced  to  its 
anthranol,  which  is  readily  reoxidized  by  exposure  to  air,  dilution 
with  water,  or,  best,  by  the  addition  of  a  few  drops  of  hydrogen 
peroxide,  a  pink-red  color  resulting.  Phthalin  is  not  sensitive  to 
this  method  of  oxidization,  and  remains  colorless.  This  procedure 
constitutes  a  much  more  specific  test  for  oxymethylanthraquinones 
than  the  Borntrager  reaction.  Warren  7  has  applied  it  to  mixtures 
of  phenolphthalein  with  senna,  cascara,  and  aloes,  and  obtained  satis- 
factory results.  In  case  of  aloes  he  found  that  upon  reoxidization 
the  red  color  was  slow  to  appear,  and  was  somewhat  masked  by 
yellow  coloring  matter.  This  point  will  be  referred  to  later.  Warren 
has  also  perfected  a  method  to  effect  the  separation  of  phenol- 
phthalein in  presence  of  oxymethylanthraquinones  which  depends 
upon  the  removal  of  phenolphthalein  as  its  tetraiodo  compound, 
leaving  the  anthraquinone  derivatives  to  be  tested  for  in  the  usual 
way. 
It  was  originally  our  hope  to  find  some  color  reaction  which 
would  serve  to  differentiate  between  the  cathartic  drugs  already 
mentioned.  While  experiments  have  thus  far  fallen  short  of  success 
in  this  respect,  they  have,  nevertheless,  led  to  some  interesting  differ- 
ences among  the  oxymethylanthraquinones  themselves,  more  espe- 
cially between  the  dioxymethyl  and  trioxymethyl  derivatives,  chryso- 
phanic acid  and  emodin  respectively. 
4Tschirch  und  Pederson,  Arch.  Pharm.,  236,  p.  205  (1898). 
5  Bailey,  /.  Ind.  Eng.  Chem.,  6,  4,  320. 
6  Of  the  two  last-named,  substances,  hematoxylin  belongs  to  a  group  of 
colors  distinct  from  the  anthracene  derivatives,  viz.,  the  flavon  group,  while 
the  constitution  of  curcumin  shows  it  to  be  closely  allied  to  the  flavon  deriva- 
tives.   Czapek,  "  Biochemie  der  Pflanzen  II,"  pp.  523  and  527. 
7  Am.  Jour.  Pharm.,  86,  10,  444-49  (1914). 
