226  PJiarmacopozial  Assays  of  Drugs  and  Galenicals.{^m- Ma/;Sarm" 
examined,  although  they  have  been  shown  to  contain  notable  quan- 
tities of  both  strychnine  and  brucine ;  they  still  await  researches 
similar  to  those  made  by  Liebig  and  others,  and  later  by  O.  Hesse, 
into  the  nature  of  the  residuary  products  of  quinine  manufacture. 
But  granting,  for  the  sake  of  argument,  that  the  two  alkaloids 
named  fairly  represent  the  total  alkaloidal  constituents,  it  has  been 
found  that  the  total  percentage  of  alkaloids  varies  in  the  commer- 
cial article  generally  between  2-5  and  3-5.  In  a  sample  of  Bombay 
seeds,  Dunstan  and  Short  determined  (Year-book,  1883,  p.  235)  3  90 
per  cent.,  and  in  one  specimen  {ibid.,  1 884,  p.  463),  taken  directly  from 
the  fruit,  5-34  per  cent,  was  obtained.  Now,  regarding  the  ordi- 
narily best  results  with  commercial  samples  (3-5  per  cent.)  as  pure 
strychnine,  one-twelfth  grain  of  this  alkaloid  would  be  represented 
by  2-38  grains  of  nux  vomica;  or  by  double  this  amount  (aY\  gr-) 
if  strychnine  be  regarded  as  constituting  one-half  of  the  total  alka- 
loids. All  these  quantities  are  within  the  limits  of  allowable  large 
doses ;  but  no  prudent  physician  would  commence  with  such  doses 
of  such  a  potent  medicine. 
There  is  still  no  process  known  by  which  strychnine  may  be 
absolutely  and  completely  separated  from  the  other  strychnos  alka- 
loids. Dragendorff  (  Werthbestimmung)  regards  the  two  principal 
alkaloids  as  being  present  in  approximately  equal  proportion. 
Dunstan  and  Short  {loc.  cit.  1883,  p.  469)  have  followed  a  method 
of  separation  which,  in  their  hands,  has  given  approximately  cor- 
rect results.  On  calculating  the  relative  percentage  of  strychnine 
to  the  total  alkaloids,  as  determined  by  them  from  commercial  tinc- 
tures and  extracts,  it  will  be  found  to  vary  for  the  tinctures  between 
32  7  and  49-8  per  cent.,  and  for  the  extracts  between  35-8  and  50-1 
per  cent.,  the  extremes  being  in  the  proportion  of  2  to  somewhat 
over  3.  It  is  known  that  brucine  has  an  action,  which  is,  qualita- 
tively, very  similar  to  that  of  strychnine,  but  quantitatively,  differs 
very  materially,  according  to  Falck,  being  weaker  in  the  proportion 
of  38-5  to  1.  Calculating,  upon  this  basis,  the  activity  of  brucine 
into  strychnine,  the  latter  would  be  represented,  instead  of  the 
mixed  alkaloids,  by  the  figures  34-5  and  51-4,  the  proportion  of 
the  lowest  and  highest,  or  weakest  and  strongest  being  very  nearly 
the  same  as  before,  2:3.  It  is  evident,  therefore,  that  the  determi- 
nation of  the  total  alkaloids  will  not  secure  the  asserted  uniformity ; 
it  will  even  not  lessen  the  uncertainty  to  any  appreciable  degree. 
