Amiilrch,i908.rm'}      Standardization  of  Preparations.  119 
the  sides  and  bottom  of  the  flask,  from  which  most  of  the  superna- 
tant liquid  can  be  decanted.  The  last  few  drops  are  evaporated  and 
the  residue  dried  to  constant  weight  at  6o°  C.  The  weight  of  digi- 
toxin  so  obtained,  multiplied  by  five,  will  express  the  percentage 
present  in  the  drug. 
The  following  table  shows  the  comparative  results  of  a  series  of 
chemical  and  physiologic  tests  on  the  same  preparations : 
Physiologic  assay  : 
Chem.  assay  :  grammes      amount  to  kill  240  gramme 
No.      Preparations  digitoxin  in  100  c.c.  pig  in  two  hours 
1.  U.  S.  P.  tincture  0*0377  o*6  c.c. 
2.  "         "   o  023  1  to  1*25  c.c. 
3.  "  "   0*0277  075  c.c. 
4.  "  "   0-0254  1  c.c. 
5.  Fat-free    "      .  0*027  1  to  1*25  c.c. 
6.  Fluidextract     .......  0*264  o"1  c-c- 
7.  "       "  0*2405  0*09  to  o*i  c.c. 
8.  "       "   0*234  0*08  c.c. 
9.  Powd.  extract  (1*061  percent.)  0*019  to  0*025  grammes. 
It  will  be  seen  from  the  table  that  quite  a  uniform  relationship 
exists  between  the  percentage  of  digitoxin  found  and  the  amount 
of  the  preparation  required  to  kill  a  standard-weight  guinea-pig  in 
a  definite  time.  Taking  any  one  of  the  preparations  in  the  table  as 
a  starting-point,  as  the  amount  of  digitoxin  in  column  two  increases, 
the  amount  of  the  preparation  required  to  kill  in  the  third  column 
decreases,  the  amounts  being  inversely  proportional.  These  results 
would  lose  much  of  their  meaning  and  value  if  the  preparations 
assayed  had  all  been  prepared  from  one  lot  of  drug.  Such,  how- 
ever, was  not  the  case,  as  the  experiment  extended  over  a  consider- 
able period  of  time  and  covered  several  different  shipments  of  digi- 
talis leaves. 
The  fat-free  tincture  referred  to  in  the  above  table  is  one  prepared 
from  drug  which  has  first  been  exhausted  with  petroleum  benzine 
to  remove  fats,  volatile  oil,  etc.,  thereby  making  the  preparation 
miscible  with  aqueous  solutions  without  causing  precipitation.  This 
treatment  has  apparently  little  or  no  effect  on  either  the  chemical 
or  physiological  assay. 
A  summary  of  my  records  on  the  assay  of  digitalis  leaves,  going 
back  about  four  years,  shows  the  following  facts  : 
Highest  percentage  of  digitoxin  0*455  per  cent. 
Lowest  "   0*171  " 
Average         "  "   0*313  '* 
