436 
PUEIFICATION  OF  BITTEE  ALMOND  OIL. 
for  that  purpose,  and  at  the  same  time  not  exert  any  action  on 
the  free  hydride  of  benzoyle.  After  many  attempts,  one  of 
which  consisted  in  trying  chloride  of  iron  and  lime,  I  finally 
adopted  the  following  plan.  I  had  found  that  on  mixing  a  small 
quantity  of  oxide  of  mercury  and  water,  and  agitating  the  oil, 
and  then  adding  a  little  caustic  potassa  and  filtering,  that  the 
oil  passed  through  clear  and  colorless  and  entirely  free  from 
cyanogen  ;  but  I  found  that  the  purified  oil  dissolves  apprecia- 
ble quantities  of  the  mercury,  and  if  chemically  pure  hydride  of 
benzole  be  digested  on  dry  oxide  of  mercury  and  filtered,  it  will 
be  found  to  dissolve  it  to  a  certain  extent.  I  therefore  pro- 
cured a  wrought  iron  still,  having  a  very  long  neck,  around  which 
a  stream  of  cold  water  was  allowed  to  pass,  so  as  to  cool  the 
vapor  sufficiently  in  order  not  to  crack  the  glass  tubes  of  the 
condenser.  I  found  that  an  ordinary  metal  worm  was  speedily 
dissolved  by  the  distillation  of  mercury.  The  pure  oil  does  not 
dissolve  metallic  mercury  as  it  does  its  oxide.  In  the  retort  of 
the  still  I  placed  a  quantity  of  crude  oil,  a  small  quantity  of 
solution  of  caustic  potassa,  and  lastly,  the  quantity  of  oxide 
necessary.  The  actual  weight  of  the  latter  has  to  be  varied. 
Some  kinds  of  oil  contain  more  cyanogen  than  others,  and  the 
requisite  quantity  of  oxide  was  determined  on  a  small  quantity 
and  filtration.  The  quantity  of  potassa  required  is,  strange  to 
say,  very  small  in  all  cases,  and  I  always  found  that,  although 
distillation  with  HgO  and  water  alone  never  succeeds  in  freeing 
the  cyanogen,  a  small  quantity  of  caustic  added  produces  per- 
fect separation.  More  alkali  must  not  be  added  than  necessa- 
ry, or  the  high  heat  of  the  distillation  will  cause  the  formation 
of  benzoate ;  and  if  the  operation  be  eff'ected  by  means  of  a  cur- 
rent of  steam,  the  water  dissolves  large  quantities  of  the  oil. 
For  this  reason  as  little  water  must  be  added  as  will  suffice  to 
dissolve  the  caustic  alkali,  and  by  careful  management  of  the 
heat  nearly  all  the  water  comes  over  before  any  oil  distils. 
Towards  the  close  of  the  operation,  large  quantities  of  a  thick 
smoke  suddenly  pour  over^  smelling  strongly  of  some  of  the 
acryle  compounds.  I  have  been  unable  to  determine  at  present 
to  what  these  fumes  are  due,  but  I  have  noticed  that  as  the  oil 
distils  it  gets  darker  in  color,  and  at  last  a  tarry  matter  contain- 
ing nitrogen  comes  over,  and  I  have  sometimes,  by  very  cautious 
