202  Arsenic  in  Bismuth  Preparations.      {Am  MayVigw?""* 
and  found  that  in  the  Marsh  test  they  easily  showed  much  smaller 
amounts  of  arsenic  than  the  maximum  allowed  in  the  Pharmacopoeia. 
I  was  confident  then  that  the  results  obtained  by  the  druggist 
were  due  to  impure  zinc  and  sulphuric  acid  used  in  the  tests  which 
he  had  carried  out ;  so  confident,  indeed,  that  I  consented,  upon  re- 
quest, to  make  a  demonstration  of  this  fact  to  the  druggist  himself. 
I  therefore  set  up  my  apparatus  in  his  laboratory  and  charged  it 
with  my  reagents,  and  showed  the  total  absence  of  arsenic  in  the 
preparation.  My  apparatus  consisted  of  the  ordinary  250  c.c. 
Florence  flask  bearing  a  doubly  bored  stopper  with  thistle  funnel 
and  delivery  tube.  To  the  latter  was  attached  a  drying  tube  fitted 
with  cotton  plugs  and  granular  calcium  chloride,  then  the  tip,  or  at 
other  times,  a  constricted  tube  for  the  deposit  of  the  arsenic 
mirror  on  heating. 
Still  the  druggist  was  not  satisfied.  He  wanted  me  to  explain 
why  he  had  gotten  the  arsenic  stain  on  porcelain  in  his  tests,  as  he, 
also,  had  used  pure  reagents.  He  then  suggested  that  I  make  a 
test  in  his  apparatus  with  my  reagents. 
His  apparatus  consisted  of  a  small  Erlenmeyer  flask  of  less  than 
100  c.c.  capacity,  bearing  a  single  bored  stopper  with  short  (3-inch) 
tube  for  the  hydrogen  tip.  On  using  his  apparatus  with  my  reagents 
I  was  surprised  to  get  a  heavy  black  deposit  (mirror)  on  the 
porcelain. 
The  druggist  seemed  satisfied  with  that  and  went  on  to  show  me 
that  the  stain  was  soluble  in  concentrated  nitric  acid  on  heating  and 
gave  with  silver  nitrate  a  reddish-brown  coloration — which  test,  he 
said,  confirmed  the  presence  of  arsenic.  (This  test  he  took  from 
Simon's  chemistry.  It  depends  upon  the  oxidation  of  arsenic  by 
nitric  acid  to  arsenic  acid  and  the  subsequent  precipitation  of  silver 
arsenate.)  I  suggested  to  him  to  try  the  more  common  test,  viz., 
the  solubility  of  the  arsenic  stain  in  sodium  hypochlorite.  This  he 
did,  but  the  stain  did  not  dissolve.  I  was  satisfied  then  that  we  were 
dealing  not  with  an  arsenic  stain,  but  with  a  bismuth  stain.  But  the 
druggist  had  never  heard  of  a  bismuth  stain,  and  neither  had  I. 
When  I  suggested  the  possibility  of  bismuth  being  carried  bodily 
over  from  his  generating  flask,  which  was  not  provided  with  any- 
thing to  arrest  it,  he  told  me  that  that  was  the  way  every  druggist 
would  make  the  test. 
I  then  asked  him  to  supply  me  with  some  French  bismuth  which 
