AmAvrn:mtm'}       Analysis  of  Polygala  Senega.  179 
ing  extract  was  light  brown  in  color,  transparent,  and  had  the 
characteristic  odor  of  the  drug.  This  residue  was  treated  with 
several  successive  portions  of  cold  distilled  water  ;  the  aqueous  solu- 
tion was  then  tested  for  tannin  with  the  following  result: 
Ferric  chloride:  brown  color,  no  precipitate. 
Bromine  water :  slight  precipitate  of  a  yellowish  color. 
Gelatin  in  presence  of  alum :  cloudiness  and  very  slight  precipi- 
tate. 
Freshly  prepared  lime  water :  yellow  color,  no  precipitate. 
Another  portion  of  the  aqueous  solution  was  shaken  out  with 
acetic  ether,  until  the  latter  remained  colorless.  The  ethereal  solu- 
tion was  evaporated  on  a  water  bath,  and  the  residue  treated  with 
cold  distilled  water ;  a  resinous  white  substance  deposited.  After 
warming  the  solution,  the  clear  supernatant  liquid  gave  the  follow- 
ing reactions  for  tannin  : 
Ferric  chloride  :  brown  color,  no  precipitate. 
Bromine  water :  yellow  color,  no  precipitate. 
Gelatin  in  presence  of  alum  :  cloudiness,  no  precipitate. 
Freshly  prepared  lime  water :  yellow  color,  no  precipitate. 
These  results  indicate  the  absence  of  tannin  in  the  drug. 
The  aqueous  extract  from  the  alcohol  residue,  after  previous 
shaking  out  with  acetic  ether,  was  precipitated  with  neutral  and 
basic  lead  acetate,  the  former  producing  a  slight,  the  latter  a  copious 
precipitate,  distinctly  yellowish  in  color.  The  glucose  present  in  the 
alcoholic  extract  amounted  to  0  08  per  cent.,  the  saccharose  to  0  50 
per  cent.  Mayer's  reagent  gave  no  precipitate.  The  portion  not 
dissolved  by  water  was  5-98  per  cent.  This  was  probably  largely 
composed  of  polygalic  acid. 
Cold  distilled  water  extracted  14-02  per  cent,  of  the  drug.  The 
infusion  presented  a  frothy  appearance.  It  contained  2  68  per  cent, 
of  glucose,  and  5-32  per  cent,  of  saccharose.  Qn  precipitating  dex- 
trin in  the  usual  way,  by  absolute  alcohol,  it  yielded  only  traces. 
An  attempt  was  made  to  remove  the  saponoid  principle  from  this 
infusion  by  means  of  several  methods. 
A  portion  of  the  dilute  infusion  was  shaken  out  with  chloroform ; 
although  caution  was  exercised  in  shaking,  it  was  almost  impossible 
to  separate  the  chloroform  in  a  perfectly  clear  state,  but  it  appeared 
frothy. 
By  slightly  warming  this  chloroformic  solution  in  a  water  bath,  it 
