474  Titration  as  a  Means  of  Assaying  Drugs.  { Am^c°t%93.arm' 
of  titration  of  alkaloidal  residues  from  assays  by  means  of  volu- 
metric acid  solution.  After  considerable  delay  the  work  has  been 
about  completed  by  both  of  us,  each  working  separately.  As  long 
as  drugs  have  been  assayed  it  has  been  customary  to  weigh  the  res- 
idue obtained  by  evaporating  the  final  extract  of  the  alkaloids  by 
ether,  chloroform  or  some  other  solvent,  and  to  call  it  alkaloid. 
This  is  frequently  accompanied  by  the  statement  that  the  alkaloids 
are  or  are  not  perfectly  pure.  How  pure  they  are  the  sequel  will  very 
plainly  show.  Beckurts,  Schweissinger  and  all  the  German  pharma- 
ceutical chemists  have  adopted  titration  with  volumetric  acid  solu- 
tion as  the  most  accurate  method  that  we  at  present  have  for  assaying 
alkaloidal  drugs,  and  there  need  be  no  reason  why  we  should  not 
adopt  it,  especially  if  the  results  of  experience  show  how  much 
nearer  the  truth  we  will  be  than  when  we  used  the  gravimetric 
method  alone.  That  this  method  is  without  blemish  we  do  not 
claim ;  in  fact,  we  are  candid  to  say  there  are  two  questionable  ele- 
ments which  enter  into  the  problem,  though  only  in  one  or  two 
instances,  and  give  rise  to  some  doubts  as  to  the  absolute  correctness 
of  our  results  in  these  instances.  Even  allowing  that  an  error  has 
been  introduced,  and  calculating  this  at  its  maximum,  we  find  that 
the  result  obtained  by  the  titration  method  is  nearer  the  truth  than 
the  result  obtained  by  the  gravimetric  method.  The  two  elements 
that  enter  the  problem  and  cause  us  to  hesitate  ere  saying  "  correct," 
in  the  cases  of  nux  vomica,  ipecac,  cinchona,  aconite  and  gelsemium, 
are :  First,  our  imperfect  knowledge  of  the  molecular  weights,  or 
rather  of  the  formulas,  of  some  of  the  alkaloids,  as,  for  instance, 
emetine,  gelsimine,  aconitine,  etc.,  and  second,  the  fact  that  some 
drugs  (nux  vomica  and  cinchona  notably),  contain  several  alkaloids 
possessing  different  molecular  weights,  and  this  compels  us  to  assume 
that  they  are  present  in  certain  proportions  in  order  to  get  the  mol- 
ecular weight  from  which  to  determine  our  percentage  of  alkaloids 
present.  The  first  difficulty  cannot  be  obviated  until  more  exact 
analyses  and  formulas  are  forthcoming,  and  confronts  us  but  seldom. 
The  second  difficulty  can  only  be  obviated  by  determining  in  each 
case  by  a  separate  assay  just  how  much  of  each  alkaloid  is  present. 
This  presents  itself  in  five  cases,  nux  vomica,  jaborandi,  veratrum 
viride,  cinchona  and  aconite.  When  we  consider  what  great  strides 
nearer  to  the  truth  we  have  taken  in  case  of  the  remaining  alkaloids 
(see  the  results  below),  and  that  we  have  in  their  cases  results  which 
