THE  AMEEIOAIf 
JOURNAL  OF  PHARMACY 
AUGUST,  igii 
THE  ESTIMATION  OF  MINUTE  QUANTITIES  OF 
NITROGLYCERIN.* 
'  By  Wilbur  L.  Scoville. 
From  a  medicinal  point  of  view  the  estimation  of  nitroglycerin 
offers  peculiar  difficulties.  This  drug  acts  powerfully  on  the  human 
system  and  is  given  in  doses  from  i/iooo  to  1/20  grain.  The  most 
common  form  of  administration  is  in  tablets,  and  a  method  of  de- 
termining the  strength  of  these  with  accuracy  is  much  to  be  desired. 
For  pharmaceutical  purposes  nitroglycerin  is  obtained  in  10  per 
cent,  alcoholic  solution,  which  is  practically  a  saturated  solution,  or 
in  about  20  per  cent,  admixture  with  absorbent  powders,  as  sugar 
of  milk,  chalk,  talcum,  etc.,  to  which  a  little  bicarbonate  of  sodium 
or  carbonate  of  magnesium  has  been  added  for  safety  in  shipping. 
There  is  evidence  that  both  the  solution  and  the  powder-mixture 
deteriorate  slowly.  In  cold  weather  a  portion  of  the  nitroglycerin 
will  separate  from  the  alcoholic  solution,  leaving  the  liquid  weak 
unless  the  precaution  is  taken  to  warm  and  redissolve.  Furthermore 
as  L.  H.  Bernegua  has  pointed  out  ^  there  is  a  loss  in  the  process 
of  manufacturing  tablets  of  nitroglycerin,  and  the  accuracy  of  the 
tablets  is  not  therefore  a  question  merely  of  mathematics  and  careful 
workmanship. 
For  the  estimation  of  nitroglycerin  in  pharmaceutical  prepara- 
tions, two  methods  are  in  general  use, — the  nitrometer  method,  and 
titration  after  saponification  with  standard  alcoholic  potash.  Of 
these  the  nitrometer  method  is  undoubtedly  the  more  accurate,  and 
*  Read  at  the  Indianapolis  meeting  of  the  American  Chemical  Society, 
July.  iQii. 
^  Amer.  Jour.  Pliar.  1907,  page  555. 
(359) 
