AmAJp°rnr;i9SrnJ-}     Contributions  from  H.  M.  Gordin.  161 
As  we  generally  do  not  know  how  much  alkaloid  is  left  behind,  it 
is  impossible  to  introduce  a  proper  correction  in  our  results. 
How  ineffective  some  of  the  proposed  methods  of  exhaustion  are 
can  be  shown  by  applying  the  Dunstan  and  Short  method  1  to  the 
assay  of  nux  vomica. 
Ten  grammes  of  dry  powdered  nux  vomica  were  extracted  in  a 
S^xhlet  with  a  mixture  of  alcohol  and  chloroform  for  three  hours, 
as  directed  by  these  authors.  The  extract  thus  obtained,  after 
passing  through  immiscible  solvents,  was  assayed  by  the  general 
N 
alkalimetric  method  above  referred  to,  using  —  acid  and  alkali  for 
40 
titration  and  Mayer's  reagent  as  precipitant.  The  dregs  were  then 
removed  from  the  Soxhlet,  dried  and  assayed  separately  by  means 
of  modified  Prollius'  fluid,  as  given  later  under  nux  vomica.  The 
final  estimation  was  again  made  by  the  above-mentioned  alkali- 
metric  method.    In  both  cases,  amounts  representing  5  grammes 
of  the  extract  and  dregs  respectively  were  taken  up  by  20  c.c. 
40 
acid.  The  factor  taken  was  the  mean  factor  of  strychnine  and 
brucine. 
EXTRACT. 
N 
—  acid  consumed  by  5  grammes,  13-5  c.c.  —  2-46  per  cent,  total 
40 
alkaloids. 
DREGS. 
N 
—  acid  consumed  by  5  grammes,  27  c.c.  =  0-49  per  cent,  total 
40 
alkaloids. 
By  taking  1  gramme  of  talcum  with  definite  amounts  of  —  acid, 
40 
making  up  the  liquid  to  100  c.c,  filtering  off  50  c.c.  and  titrating 
N 
with —  alkali,  using  phenolphtalein  as  indicator,  I  find  that  this 
40 
amount  of  talcum  consumes  2  c.c.  —  acid     It  is  therefore  best  to 
40 
use  1  gramme  talcum  in  all  cases  where  the  precipitate  retuses  to 
separate  out,  and  deduct  2  c.c.  from  the  total  amount  ol  ?  acid 
40 
consumed  by  the  alkaloid. 
^rescott's  "  Organic  Analysis,"  1887,  456. 
