Am/X?i?86arm'}  Estimation  of  Cocaine  by  Mayer's  Reagent.  277 
for  estimation  of  cocaine  are  lacking  in  exactness.  A  variation, 
under  conditions  as  nearly  identical  as  are  likely  to  be  secured  in 
ordinary  practice,  amounting  to  10,  and  even  15  per  cent,  of  the 
whole  quantity  of  alkaloid  present  is  not  uncommon,  where  the  solu- 
tion titrated  is  too  much  concentrated.  Still  it  may  be  that  an  attempt 
to  separate  the  alkaloid  by  other  assay  processes,  such  as  are  com- 
monly adopted,  will  involve  even  greater  possibilities  of  error,  ex- 
cept in  experienced  hands,  and  a  result  which  may  be  relied  upon 
within  even  so  wide  a  range  has  positive  value.  A  more  important 
question  is :  Can  the  titration  method  be  applied  to  the  fluid  extract, 
tincture,  etc.,  of  coca  leaves  so  as  to  yield  useful  results  ?  This  in- 
volves the  question  whether  there  is  present  in  the  drug  more  than 
one  alkaloid,  and  if  so,  whether  the  alkaloids  differ  from  one  another 
greatly,  (a)  in  physiological  action,  and  (b)  in  their  deportment  with 
Mayer's  reagent.  We  are  not  yet  sufficiently  familiar  with  the 
chemistry  of  the  cocaine  alkaloids  to  answer  these  questions.  I  find 
that  the  quantity  of  alkaloid  indicated  by  Mayer's  reagent  in  coca 
leaves  and  their  preparations  is  considerably  greater  than  the  quantity 
of  cocaine  I  have  been  able  to  extract  by  washing  out  the  alkaline 
solution  with  ether.  Chloroform  will  extract  an  additional  amount 
of  alkaloid,  but  the  total  alkaloid  extracted  falls  short  of  what  is 
indicated  by  Mayer's  reagent — generally  in  the  neighborhood  of 
two  per  cent.,  estimated  as  cocaine.  The  cause  of  the  discrepancy 
calls  for  investigation  which  I  have  not  been  able  to  give  to  it.  As 
long  as  the  subject  is  so  imperfectly  understood,  I  prefer  to  depend 
on  actual  weighings  of  alkaloid,  but  the  separation  from  a  complex 
fluid  of  an  alkaloid  so  sensitive  to  the  action  of  acids  and  alkalies  is 
a  matter  of  no  little  difficulty,  and  I  am  inclined  to  believe  that 
experience  will  show  that  the  comparative  value  of  fluid  extracts,  for 
example,  of  coca  leaves — can  be  practically  ascertained  by  the  use  of 
Mayer's  reagent. 
Instead  of  depending  on  titration,  in  case  we  have  to  deal  with 
solutions  of  the  alkaloid,  we  may  collect  the  precipitate  on  a  double 
filter,  wash  slightly,  in  such  a  way  as  to  wash  the  filters  more  than 
the  precipitate,  dry  and  weigh.  The  weight  of  the  precipitate  multi- 
plied by  0*406  will  give,  quite  closely,  the  weight  of  alkaloid  present. 
If  titration  is  preferred,  and.  Mayer's  reagent  of  one-half  strength 
is  used,  we  may  adopt  the  following  values  for  the  equivalent  of  the 
reagent : 
