160  Oil  of  Bitter  Almonds.  { Am  imrm' 
character  appeared  in  the  Journal  of  the  Society  of  Chemical  Industry, 
1904,  303,  wherein  he  mentions  an  attempt  to  apply  this  reaction 
to  the  determination  of  benzaldehyde  in  oil  of  bitter  almonds,  and 
states  that  they  gave  "  results  varying  about  one-third  of  the  theo- 
retical.'' Sadtler's  latest  contribution  on  this  subject  [Jour.  Amer. 
Chem  Soc,  1905,  1 3 2 1)  deals  with  the  benzaldehyde  determination 
slightly  more  in  extenso.  He  gives  results  obtained  in  the  estima- 
tion of  benzaldehyde  in  a  good  quality  of  the  commercial  article, 
which  ranged  on  the  same  sample  from  96  8  per  cent,  to  99  percent. 
No  other  analytical  results  are  recorded,  but  we  are  led  to  conclude 
that  the  method  works  well,  although  the  operation  must  be  carried 
out  entirely  in  the  cold,  the  avoidance  of  heat  being  absolutely 
necessary.  It  is  from  this  work  that  the  U.S.P.  method  of  assay  is 
probably  derived. 
Judging  by  the  results  below,  it  seems  that  the  introduction  of 
this  method  of  assay  into  the  Pharmacopoeia  has  been  a  little  pre- 
mature, for  I  have  been  unable  to  obtain  results  that  are  reliable, 
and  it  further  appears  that  the  same  conclusions  regarding  this 
matter  have  been  reached  by  the  chemists  of  Schimmel  &  Co. 
They  say  {Schimmel 's  Report,  1905,  30)  "  we  were  compelled  to 
reject  the  use  of  Sadtler's  method  for  the  estimation  of  citral  in 
lemon  oil  because  an  exact  titration  could  not  be  carried  out,"  and 
again  {SchimmeVs  Report,  April-May,  1906,  70,  74  and  122)  they 
reiterate  this  opinion.  This  is  exactly  the  difficulty  which  I  have 
encountered  as  being  the  most  serious  one. 
The  application  of  this  assay  to  sample  number  1,  which  was  a 
sample  of  the  genuine  oil  from  Fritzsche  Brothers,  and  which  an- 
swered all  the  other  U.S.P.  tests,  gave  the  following  results  : 
69-  i9  per  cent  ,  .      .  .  68'o6  per  cent. 
70-  5    "       "   69  46    "  " 
68  93  "       "      ...........  69-34  "  " 
The  average  of  these  results  is  69  25  per  cent.,  and  the  extreme 
variation  among  these  is  2- 54  per  cent.,  which,  when  we  consider 
the  method  of  assay,  is  sufficiently  close  agreement  for  all  practical 
purposes.  In  view  of  the  requirement  of  the  U.S.P.  that  oil  of 
bitter  almonds  should  contain  85  per  cent,  of  benzaldehyde,  these 
results  may  be  looked  upon  with  suspicion  as  showing  not  a  poor 
quality  of  oil,  but  the  shortcomings  of  the  process.  The  quantity 
of  benzoic  acid  present,  as  shown  by  two  assays,  was  0  95  and  0  98 
