American  Pharmaceutical  Association. 
483 
were  obtained  by  the  authors  last  year  seemed  to  indicate  that  kolatannin  was 
a  glucosidal  body;  but  after  repeated  experiments,  in  which  the  glucose  of  the 
plaut  was  entirely  eliminated  from  the  tannin,  negative  results  were  obtained 
for  glucose  resulting  from  decomposition  of  the  kolatannin,  and  the  inter- 
pretation was  made  that  kolatannin  is  not  a  glucoside.  The  authors  said  it  is 
significant  that  this  result  places  kolatannin  in  a  category  and  gives  it  a  char- 
acter essentially  different  from  the  character  of  caffetannic  acid  as  understood 
by  the  chemical  world  from  the  report  or  Hlasiwetz.  In  every  case  kolatannin 
yields  at  once  a  pentacetyl  derivative  ;  it  readily  admits  successively  three, 
four  and  five  atoms  of  bromine  in  substitution  for  hydrogen,  and  these  several 
bromine  and  acetyl  substitutions  are  obtained  just  the  same,  whether  the 
bromine  substitution  or  the  acetyl  substitution  be  made  first  in  order.  A  sixth 
atom  of  bromine  can,  with  more  difficulty,  be  introduced,  but  this  product  will 
at  most  receive  only  four  groups  of  acetyl,  forming  tetracetylhexabromkola- 
tannin.  In  analysis  of  the  acetyl  derivatives,  the  figures  from  combustion  were 
confirmed  by  a  method  of  saponification. 
Kolatannin  itself  forms  a  first,  a  third  and  a  fourth  anhydride,  and,  like 
anhydrides,  are  formed  by  the  several  bromine  derivatives,  all  these  purifying 
intact,  so  as  to  give,  on  combustion,  figures  quite  close  to  the  theoretical  ones. 
By  boiling  kolatannin  with  dilute  acids,  an  insoluble  "red"  is  obtained,  in 
place  of  the  "kola-red"  of  Knebel,  and  corresponding  to  the  oak-red,  and  the 
phlobaphene  of  Grabowski,  Etti  and  others,  but  with  utmost  efforts  this  body 
could  not  be  prepared  of  constant  composition. 
The  figures  from  six  preparations  ranged  C,  50*45,  53*60,  56*38  per  cent., 
etc.  The  authors  believe  it  to  be  a  mixture  of  two  or  more  compounds.  The 
report  of  Knebel  and  of  Hilger  that  sugar  is  formed  coincident  with  the  red 
body  was  not  confirmed  by  the  present  experiments.  Kolatannin,  under  action 
of  fused  potassium  hydroxide,  was  found  to  yield  protocatechuic  acid,  as  well 
as  phloroglucin,  indicating  both  di-  and  trihydroxy  benzoic  constitution.  Pro- 
tocatechuic acid  was  also  obtained  by  heating  with  glycerine.  The  authors 
stated  their  desire  to  make  further  determinations  of  decomposition  products  of 
kolatannin,  and  proposed,  for  further  studies,  the  following  constitutional  for- 
mula, which  they  think  is  consistent  with  the  results  so  far  obtained  : 
The  authors  also  discussed  the  two  processes  for  the  assay  of  kola  which 
have  been  proposed  in  the  past  year  by  Jean  and  Carles.  They  gave  the  essen- 
tial points  of  these  methods  as  follows  :  Method  of  Jean. — This  author  boils  the 
dried  and  powdered  drug  with  milk  of  lime,  dries  the  whole  in  an  oven  and 
powders  it  again.  This  powder  is  exhausted  with  chloroform,  which  removes 
the  free  alkaloids.  The  chloroform  is  evaporated  to  dryness,  the  residue  dis- 
solved in  hot  water  and  filtered.  The  filtrate  is  evaporated  to  dryness  and 
weighed  as  caffein.  For  "kolanin:"  the  drug,  after  treatment  with  chloro- 
form, is  extracted  with  alcohol,  the  alcohol  evaporated  from  the  percolate,  and 
the  soft  extract  remaining  is  dissolved  in  boiling  water,  which  solution,  after 
cooling,  is  filtered.  The  insoluble  matter  remaining  in  the  filter  is  dried  in  an 
oven  and  weighed  as  "  kolanin." 
The  authors,  in  criticising  the  foregoing  methods,  state  that  the  most  serious 
C6H3(CH3)(OH)3CO 
C6H3<CH3)(OCH3)(OH)20 
—  Qe-^  20*^8 
