36 
Presence  of  Arsenic  in  Chemicals. 
( Am.  Jour.  Pharm, 
I     January.  1903. 
8.  — 0-00781  1/8298  =0-00012  Stain  quite    Foil  but  little    Results  nega- 
marked  in    colored.    Crys-  tive. 
one-half   tals  plain. 
hour. 
9.  — 0*00390  1/16596=0  "00006  Stainless    Foil     very    Results  nega- 
marked  than    slightly  col-  tive. 
No.  8.  ored.  Very  few 
crystals. 
10.  — 0-00195   1/33333=0*00003  Stain  faint  in    Slight  stain.    Results  nega- 
o  n  e  -  h  a  1  f   No  crystals.  tive. 
hour. 
11.  — 0-00129    1/50000=0*00002  Stain  very    Slight  stain.        Results  nega- 
faint  in  one  tive. 
hour. 
These  figures  are  only  approximately  equivalent. 
According  to  the  above  results  the  limit  of  the  BettendorfT  test 
is  about  1 -30  milligramme  or  1-2000  of  a  grain  of  arsenous  oxide 
per  cubic  centimetre  of  solution  or  gramme  of  material.  The  limit 
of  the  Reinsch's  method  is  reached  at  about  1-260  of  a  milli- 
gramme or  I- 1 7000  of  a  grain  of  arsenous  oxide  in  1  c.c.  of  solution 
or  1  gramme  of  material.  The  limit  of  the  Marsh-Berzelius  test 
is  reached  at  about  1-512  of  a  milligramme  or  1-33333  of  a  grain 
of  arsenous  acid  per  cubic  centimeter  of  solution  or  gramme  of 
material. 
It  is  claimed  by  some  observers  that  the  i-iooo  of  a  milligramme 
per  cubic  centimetre  of  liquid  gives  positive  indications,  but  some 
of  the  more  conservative  are  of  the  opinion  that  1-10  of  a  milli- 
gramme per  cubic  centimetre  of  fluid  is  about  the  limit.  The  evi- 
dence of  the  presence  of  1-512  of  a  milligramme  of  arsenous  oxide 
per  gramme  of  material  obtained  by  the  writer  is  not  at  all  positive, 
and  he  is  of  the  opinion  that  considerably  more  must  be  present 
before  the  chemist  can  make  a  positive  statement;  i-io  of  a  milli- 
gramme is,  however,  a  little  too  conservative,  because  with  this 
amount  the  tube  becomes  almost  black  throughout  the  constriction. 
It  was  decided  to  make  standard  tubes  or  mirrors  for  the  Marsh- 
Berzelius  test  *  by  depositing  the  metallic  arsenic  in  the  tubes 
and  using  them  for  comparison  in  deciding  as  to  how  much  arsenic 
a  given  substance  contained.  This  was  decided  on,  because  it  was 
soon  found,  after  a  little  work  was  done,  that  it  was  practically 
impossible  to  weigh  the  small  quantities  of  arsenic  generally 
obtained.  The  figures  given  above  represent  the  amount  of  arsenous 
oxide  contained  in  one  or  more  cubic  centimetres  of  solution  of  known 
