442 
Ge hemic  Acid. 
f  Am.  Jour.  Pharm. 
\  September,  1897. 
The  dry  sodium  salt  (so-called)  when  heated  becomes  very  volu- 
minous, a  phenomenon  similar  to  the  "  Pharaoh's  Serpent "  pro- 
duced on  heating  mercury  sulphocyanide. 
I  next  attempted  to  produce  a  salt  with  the  alkaline  earths/by 
boiling  gelsemic  acid  with  freshly-precipitated  barium  carbonate  and 
water,  also  magnesium  carbonate  and  water  for  several  hours.  The 
filtered  solution  was  neutral,  but  upon  concentrating,  the  carbonated 
alkaline  earth  gradually  separated  and  the  solution  assumed  an  acid 
reaction.    No  crystals  separated  from  the  solution  upon  standing. 
From  the  above  it  will  be  seen  that  this  principle  possesses  very 
feeble  acid  properties  and  that  its  compounds  are  of  an  exceeding 
unstable  character. 
Attempts  were  made  to  produce  salts  by  double  decomposition 
between  the  sodium  compound  of  G.  A.  in  solution,  with  salts  of  the 
metals,  but  the  precipitates  obtained  were  of  the  same  character  as 
those  mentioned  under  test  4. 
Lassaigne's  test  for  the  presence  of  nitrogen  was  made  with  nega- 
tive results,  confirming  Robbins'  test. 
Robbins,  after  making  two  combustions  of  gelsemic  acid  with  cop- 
per oxide  in  a  simple  bayonet  tube,  as  was  customary  at  that  time, 
and  comparing  his  results  with  the  older  aesculin  formula  of  Rochle- 
der,  comes  to  the  conclusion  that  his  gelsemic  acid  is  identical  with 
aesculin,  reinforcing  his  opinion  by  comparing  the  fluorescent  prop- 
erties of  both  and  their  reducing  powers  on  Fehling's  solution.  It 
is  true  that  aesculin  and  gelsemic  acid  resemble  each  other  in  some 
particulars,  such  as  fluorescence  and  reducing  powers,  but,  as  will  be 
shown  later,  it  will  be  seen  that  the  latter  is  a  distinctively  different 
principle. 
Tne  two  combustions  of  Robbins  resulted  as  follows  : 
I.  C  =  52  04  per  cent.    H  =  5-189  per  cent. 
II.  C  ±=  51-82  per  cent.    H  =a±  4-98  per  cent. 
The  older  formula  of  Rochleder  for  aesculin  contains  C  51*57  Per 
cent,  and  H  4-87  per  cent. 
The  later  accepted  formula  contains  C  52-94  per  cent,  and  H 
4-70  per  cent. 
The  results  of  Robbins' analyses  and  the  above  formula  corre- 
spond quite  closely.  However,  the  author  questions  the  accuracy  of 
the  (Robbins)  analyses  and  the  formula  deduced  therefrom. 
The  greatest  difficulty  was  experienced  in  obtaining  concordant 
