Am.  Jour.  Pharni.) 
March,  1894.  j 
Arsenic  in  Glycerin. 
147 
of  arsenic.  The  sample  No.  1  was  of  a  character  not  likely  to  be 
met  with  in  ordinary  retail  trade,  being,  in  fact,  the  crude  material 
used  in  the  manufacture  of  nitroglycerin.  The  largest  quantity 
found  in  samples  Nos.  2  and  8  amounted  to  only  one  grain  of 
metallic  arsenic  in  fourteen  pounds  of  glycerin. 
It  appears,  therefore,  that  with  the  exercise  of  due  care  in  apply- 
ing the  test  above  described,  there  is  really  no  difficulty  in  obtaining 
glycerin  practically  free  from  dangerous  contamination  with  arsenic. 
But  it  is  desirable  that  this  point  should  be  determined  by  applying 
the  test  in  all  cases. 
The  ordinary  form  of  Marsh's  test  is  inapplicable  for  the  detec- 
tion of  minute  quantities  of  arsenic,  and  with  all  but  one  of  the 
samples  examined  we  failed  to  obtain  satisfactory  indications  of  the 
presence  of  arsenic  in  the  glycerin,  which  was  subsequently  found  to 
contain  it. 
The  method  of  testing  employed  for  this  purpose  was  a  modifica- 
tion of  that  introduced  by  Gutzeit,  which  consists  in  allowing  the 
gas  slowly  evolved  from  sulphuric  acid  and  zinc,  in  contact  with  the 
liquid  supposed  to  contain  arsenic,  to  act  upon  filter  paper  moist- 
ened with  a  saturated  solution  of  argentic  nitrate  and  placed  over  the 
mouth  of  the  test  tube  in  which  the  reaction  takes  place.  If  arsenic 
is  present,  arsenietted  hydrogen  is  formed,  and  a  yellow  stain  is 
produced  upon  the  filter  paper.  For  application  to  glycerin  this 
test  has  been  modified  by  Vulpius,  Fliickiger  and  Siebold.  Hydro- 
chloric acid  is  substituted  in  place  of  sulphuric,  and  mercuric 
chloride  is  used  instead  of  argentic  nitrate.  A  mixture  of  two 
cubic  centimetres  of  the  glycerin  to  be  tested  with  5  cc.  of  hydro- 
chloric acid  (1  to  7)  and  1  gramme  of  pure  zinc  is  placed  in  a  long 
test  tube,  the  mouth  of  which  is  covered  with  a  disc  of  filter  paper 
previously  moistened  with  one  or  two  drops  of  mercuric  chloride 
solution  and  then  dried.  If  arsenic  be  present  in  any  considerable 
amount  a  yellow  stain  is  produced  upon  the  filter  paper  after  a  few 
minutes,  and  it  subsequently  becomes  darker.  When  the  amount 
of  arsenic  is  very  minute  it  is  necessary  to  allow  the  action  to  con- 
tinue for  a  longer  time.  Fifteen  minutes  is  considered  by  Mr.  Sie- 
bold sufficient  for  practical  purposes,  and  if  no  stain  is  produced  in 
that  time  the  glycerin  may  be  regarded  as  free  from  arsenic.  But 
though  extremely  small  quantities,  such  as  Tfa  or  y^Vo  of  a  milli- 
gramme of  arsenic,  are  indicated  by  this  test,  we  think  it  is  prefer- 
able to  allow  a  longer  time  to  elapse  in  testing. 
