Am.  Tour.  Pharm.  j 
May,  1919.  J 
Preparation  of  Arsphenamine. 
305 
arsphenamine  is  ground  and  further  dried'  until  of  constant  weight. 
The  yield  is  about  75  per  cent. 
III.  The  Product. — Color. — The  substance  prepared  by  the 
method  just  described  is  usually  a  greyish  powder  sometimes  having 
a  slight  yellowish  tint.  When  very  pure  it  seems  to  have  little  or 
no  color  and  is  practically  white  in  a  dry  state,  and  is  possibly  analo- 
gous to  the  color  changes  of  cupric  sulphate  when  made  anhydrous. 
Analysis. — Analyses  of  preparations  having  about  1  molecule 
of  water  in  the  final  product  are  as  follows : 
Preparation. 
Per  Cent.  As.13 
Per  Cent.  N. 
Per  Cent.  CI. 
Found. 
Theory. 
Found. 
Theory. 
Found. 
Theory. 
No.  2  
32.67 
32.85 
No'.  5  
14.89 
15-50 
No.  12  
32.89 
32.85 
6-37 
6.13 
14.09 
15-50 
Preparations  with  2  molecules  of  water  in  the  final  product  gave 
Preparation. 
Per  Cent.  As.  13 
Per  Cent.  N. 
Per  Cent.  CI. 
Found. 
Theory. 
Found. 
Theory. 
Found. 
Theory. 
No.  9  
31-83 
31-57 
5.60 
5-90 
16.79 
14.90 
No.  10  
31-83 
31-57 
5-64 
5-90 
14.09 
14.90 
No.  11  
30.38 
31-57 
5-83 
5-90 
14.67 
14.90 
The  analyses  that  Ehrlich  and  Bertheim  obtained  on  their 
product  are  given  below  for  the  purpose  of  comparison.  One 
analysis  made  by  us  of  arsenobenzene  is  also  given.  The  theoretical 
values  are  calculated  on  the  basis  of  I  molecule  of  methyl  alcohol  to 
1  molecule  of  arsphenamine. 
Preparation. 
Per  Cent.  As. 
Per  Cent.  N 
Per  Cent.  CI. 
Found. 
Theory. 
Found. 
Theory. 
Found. 
Theory. 
Ehrlich  and  Bertheim's  .  .  . 
Arsenobenzol  No.  1449  .  .  . 
31-99 
30.89 
31-85 
31-85 
6.06 
6.20 
5-95 
5-95 
14-51 
15.24 
15-07 
15-07 
13  The  above  results  are  in  most  cases  the  averages  of  the  following  anal- 
yses. The  arsenic  estimations  were  made  according  to  the  gravimetric 
method  described  in  Treadwell-Hall  (1914)  and  also  in  Public  Health  Re- 
ports, 33,  1003-1018,  1918.  The  nitrogen  was  determined  according  to  the 
Kjeldahl  method  (modified  Gunning),  U.  S.  Department  of  Agriculture, 
Bull.  107,  8d,  1907.  The  chlorine  was  determined  by  fusing  the  product  with 
two  parts  of  potassium  nitrate  and  one  part  of  sodium  carbonate  and  then 
proceeding  according  to  Drechel  modification  of  Volhard's  method.  Tread- 
well-Hall, 2,  707,  1914. 
