2IO 
Keeping  Properties  of  Digitalis. 
(Am.  Jour.  Pbarm. 
\       May,  1913. 
per  kilo  of  frog,  which  is  about  25  per  cent,  higher  than  the  average 
as  determined  by  Hale,  and  by  Famulener  and  Lyons.10 
It  is  probable,  however,  that  the  results  obtained  by  the  cat 
method  are  the  more  nearly  correct  in  this  case,  for  it  is  well  known 
that  frogs  vary  considerably  in  susceptibility  to  the  digitalis  bodies, 
such  differences  have  been  discussed  fully  in  the  article  previously 
cited,9  and  we  would  refer  the  reader  to  that  for  confirmation  of 
the  statement. 
Turning  to  the  fluid  preparations,  we  found  that  a  sample  of 
the  fluid  extract  made  over  ten  years  ago  gave  a  cat  unit  of  no 
milligrams  of  leaf  per  kilo.  This  specimen  was  made  with  50  per 
cent,  alcohol  as  the  menstruum,  and  probably  showed  no  dete- 
rioration. 
A  sample  of  fluid  extract  of  digitalis  which  was  said  to  be 
"  not  less  than  thirty  years  old  "  was  then  tested  on  the  cat,  three 
tests  giving  units  of  130,  162,  and  153  milligrams  per  kilo  respec- 
tively, an  average  cat  unit  of  148  milligrams,  the  action  being  per- 
fectly typical  of  digitalis.  As  we  have  no  means  of  knowing  the 
original  activity  of  the  leaf  from  which  this  fluid  extract  was  made 
we  might  assume  that  it  was  of  the  average  strength,  that  is,  that 
it  would  originally  have  shown  a  unit  of  about  100  milligrams. 
On  this  basis  we  might  suppose  that  in  more  than  thirty  years  it 
had  declined  only  about  40  per  cent,  in  activity.  As  a  matter  of 
fact,  it  was  more  active  by  32  per  cent,  than  the  average  of  thirteen 
specimens  of  fluid  extract  obtained  in  commerce  in  the  present 
year,  the  explanation  being  that  it  is  especially  difficult  to  prepare 
a  fluid  extract  of  digitalis  which  represents  the  full  activity  of  the 
leaf. 
This  thirty-year-old  fluid  extract  having  been  made  according 
to  the  Pharmacopoeia  of  1870,  had  a  menstruum  composed  of  about 
70  per  cent,  alcohol,  20  per  cent,  glycerin,  and  10  per  cent,  water. 
Tests  of  this  specimen  by  the  one-hour  frog  method  gave  a  fatal 
dose  of  about  1300  milligrams  per  kilo  of  frog.  This  is  almost 
certainly  too  high  a  figure,  and  may  be  attributable  to  the  presence 
of  glycerin  in  the  preparation.  Glycerin  often  delays  absorption 
from  the  lymph-sac  of  the  frog  and  makes  the  specimen  which  con- 
tains it  seem  weaker  than  it  actually  is,9  but  this  is  without  influence 
in  the  case  of  tests  made  on  the  cat  by  our  method. 
10  Proc.  Am,.  Pharm.  Association,  L,  1902,  p.  415. 
