238 
Tannin  in  Quercus  Alba. 
Am.  Jour.  Pharm. 
May,  1890. 
variation,  but  the  two  fractions  of  which  duplicates  were  made  and 
which  agreed  most  closely,  gave  of 
These  results  correspond  approximately  to  the  formula  C29H27013. 
Owing  to  this  instability  of  the  tannin  of  Quercus  alba,  it  will  be 
apparent  that  the  results  of  combustions  must  vary  and  be  rather 
unsatisfactory.  In  arriving  at  the  constitution  of  this  tannin,  the 
solution  probably  lies  in  preparing  acetyl  or  benzoyl  derivatives 
and  other  stable  compounds  with  it,  some  of  which  I  hope  shortly 
to  be  able  to  prepare. 
The  derivatives  of  this  tannin  are  interesting,  as  they  are  not 
pyrogallic  acid  or  phloroglucin,  and  hardly  pyrocatechin  or  protoca- 
techuic  acid.  A  portion  of  the  tannin  was  heated  between  two 
watch  crystals,  when  a  carbonized  and  tasteless  mass  remained  ; 
while  upon  the  upper  crystal  there  sublimed  in  needle-like  crystals  a 
light  yellow  compound,  but  sparingly  soluble  in  water,  the  solution 
having  a  blue  fluorescence  ;  readily  soluble  in  alcohol  and  in  potas- 
sium hydrate,  producing  with  the  latter  a  red  color  which  possessed 
an  evanescent  blue  fluorescence.  It  resembled  pyrocatechuic  acid, 
in  that  it  gave  a  dark  green  color  with  ferric  chloride,  which  is 
turned  slightly  red  by  potassium  hydrate  and  by  hydrochloric  acid 
the  green  is  restored.  It  differs  from  this  acid,  in  that  it  does  not 
give  a  violet  color  to  fir-wood,  moistened  with  hydrochloric  acid; 
also  in  that  with  calcium  hydrate  it  produces  a  yellow  color  by 
transmitted  light  and  a  blue  color  by  reflected  light.  In  nitric  acid 
it  dissolves  with  a  yellowish-red  color,  becoming  deep  red  on  addi- 
tion of  potassium  hydrate. 
Another  portion  was  heated  with  nearly  an  equal  weight  of 
potassium  hydrate  in  a  silver  crucible  to  a  uniform  state  of  fusion. 
The  residue  was  dissolved  in  water,  acidified  with  hydrochloric  acid 
and  then  shaken  with  ether  and  the  ethereal  solution  evaporated 
nearly  to  dryness.  The  result  was  a  reddish-brown  amorphous  sub- 
stance, producing  a  dirty  mixture  with  water,  and  upon  the  addition 
of  alkalies  changing  to  a  clear  red  solution,  with  a  slight  blue  fluo- 
rescence. With  ferric  chloride  it  produced  like  protocatechuic  acid 
a  dark  green  color,  which,  upon  addition  of  potassium  hydrate,  was 
immediately  changed  to  a  bright  red  color,  but  differed  from  this 
c 
H, 
O, 
5874 
4'5o 
3676 
59^5 
4-65 
357o 
