430 
Stability  of  Digitalis  Leaf  Extracts.  { Am-j]°iyT\^grm- 
inactive.  Kraft  hydrolysed  this  saponin  and  detected  by  their  osazones,  glu- 
cose and  arabinose  (or  xylose). 
Kraft  then  reextracted  his  leaf  with  50  per  cent,  alcohol  and  on  remov- 
ing the  digitoxin  as  above  described  found  that  a  hot  alcohol  extraction  sub- 
sequent to  the  chloroform  extract  yielded  "gitin"  a  crystalline  galactoside, 
insoluble  in  water  and  in  chloroform,  melting  at  260-2650.  It  crystallizes  in 
needles  and  contains  11. 5  per  cent,  water  of  crystallization. 
The  experimental  work  reviewed  agrees  in  that  a  50  per  cent, 
alcoholic  extract  is  found  to  yield  more  active  extracts  than  a  pre- 
liminary water  extract.  For  example,  Kiliani  and  Windaus  made 
an  infusion  (10  Kg.  leaf),  from  which  190  Gm.  of  tannate  was  pre- 
cipitated, while  a  subsequent  50  per  cent,  alcohol  extraction  yielded 
after  removing  certain  fractions  with  absolute  alcohol,  chloroform 
and  ether,  150  Gm.  of  very  active  water  soluble  material  containing 
"  digitalein."  This  matter  is  very  puzzling  in  view  of  the  fact  that 
we  consider  a  fresh  10  per  cent,  infusion  as  practically  equivalent 
to  a  10  per  cent,  tincture  when  examined  by  the  one-hour  frog 
method.  Work  on  this  point,  and  on  the  rate  of  deterioration  of 
infusions  is  in  progress  in  this  laboratory.  Focke14  states  that  the 
water  extracts  about  80  per  cent,  of  the  activity.  Concerning  the 
rate  of  deterioration  of  the  infusion,  results  are  very  contradictory. 
Several  recent  observations15  have  indicated  that  alcoholic  ex- 
tracts are  not  stable.  This  finding  is  of  the  utmost  importance  in 
therapy,  and  a  clearer  understanding  of  these  values  is  important 
with  reference  to  the  chemical  investigation  of  digitalis  leaf.  One  of 
the  most  interesting  points  in  this  connection  would  be  to  observe 
whether  or  not  these  solutions  finally  arrived  at  a  permanent  equi- 
librium as  to  toxicity,  and  to  observe  what  part  of  the  toxicity  was 
thus  found  to  be  stable.  A  chemical  study  of  a  digitalis  extract 
aged  in  this  manner  would  certainly  be  a  more  hopeful  one  to  begin 
with.  Is,  for  example,  the  permanent  toxicity  of  an  aged  tincture 
the  water-soluble  fraction  or  is  it  the  "  digitoxin  "  ?  Or  do  we  have 
a  mixture  of  cleavage  products?  A  series  of  examinations  on  sam- 
ples of  alcoholic  extracts  which  we  report  below  prove  that  a  con- 
dition of  greater  stability  is  obtained  on  aging.  These  results  are 
discussed  in  the  summary  at  the  end  of  this  paper. 
14  Arch.  Pharm.,  249,  323  (1911). 
15  Roth,  Bull.  Hygienic  Lab.,  102  (1916)  ;  Pittenger  &  Mulford,  J.  Am. 
Pharm.  A.,  7,  236  (1918). 
