364  Estimation  of  Nitroglycerin.         { '^'";ug°"^;  m?'"" 
nitroglycerin  (calculated)  was  measured  into  a  small  porcelain 
evaporating  dish,  and  allowed  to  evaporate  spontaneously.  Into 
another  dish  was  measured  1.2  c.c.  of  the  standard  nitrate  solution 
and  evaporated  at  a  low  temperature.  When  both  were  dry  2  c.c. 
of  the  phenoldisulphonic  acid  reagent  were  added  to  each,  the  mixt- 
ure stirred  well  with  a  glass  rod  and  allowed  to  stand  10  minutes, 
then  diluted  with  water,  rendered  slightly  alkaline  with  potassium 
hydroxide,  cooled  and  diluted  to  100  c.c. — or  200  mm.  in  the  com- 
parison tubes.  The  colors  were  then  compared  in  a  Schreiner  colori- 
meter in  the  usual  way. 
If  a  colorimeter  is  not  at  hand,  Nessler  tubes  will  give  very 
good  satisfaction.  In  the  colorimeter  a  difference  of  ^/oq  or  5  per 
cent,  is  easily  discerned. 
For  tablets,  five  ^/^oo  grain  tablets  are  powdered,  to  c.c.  of  alcohol 
added  and  the  mixture  shaken  frequently  during  i  to  2  hours,  then 
filtered. 
Two  c.c.  of  the  clear  filtrate  is  then  evaporated  and  treated. 
Other  strength  tablets  are  treated  similarly,  the  equivalent  of  ^/loo 
grain  being  taken  for  test.  If  the  tablets  are  easily  friable  they  are 
broken  up  with  a  glass  rod  after  adding  the  alcohol. 
The  test,  so  far  as  tablets  are  concerned  was  proven  by  adding  a 
known  alcoholic  solution  of  nitroglycerin  to  varying  quantities  of 
sugar  of  milk,  drying  without  heat  then  treating  with  alcohol  as 
above  and  applying  the  test.  The  full  amount  of  nitroglycerin  put 
in  was  recovered  by  the  test,  except  in  one  instance  when  an  exces- 
sive amount  of  sugar  of  milk  was  used. 
A  sample  of  the  U.S. P.  Spirit  of  Nitroglycerin  containing  i  per 
cent,  by  weight  of  nitroglycerin  was  made  from'  solution  No.  3. 
The  specific  gravity  of  this  at  25°  C.  is  0.81378.  Ten  c.c.  of  it 
made  a  clear  mixture  with  12  c.c.  of  water  at  15°  C,  but  13  c.c.  at 
this  temperature  produced  a  marked  milkiness. 
The  present  U.S. P.  tests  on  this  spirit  allow  so  wide  a  range 
as  to  be  of  little  value.  The  above  sample  has  been  kept  six  months 
and  shows  no  change  in  strength. 
