Aocfo°be^8r6!nJ-}     Analysis  of  the  Canada  Thistle.  531 
was  poured  upon  a  filter,  the  filtrate  transferred  to  a  separatory 
funnel  and  shaken  with  several  successive  portions  of  ether,  and 
then  with  chloroform.  The  solution  was  then  made  alkaline  with 
sodium  hydrate,  and  again  treated  with  chloroform  and  ether.  The 
four  portions  of  these  solvents  which  had  thus  been  applied  to  the 
watery  liquid  were  first  heated  to  recover  the  greater  part  of  them, 
and  afterwards  allowed  to  evaporate  spontaneously.  Each  of  the 
residues  so  obtained  was  treated  with  a  small  quantity  of  warm 
absolute  alcohol.  These  solutions  were  filtered  and  the  alcohol 
evaporated.  The  residue  that  represented  the  substances  removed 
from  the  aqueous  liquid,  while  acid,  by  agitation  with  ether,  was 
crystalline,  while  the  one  obtained  with  chloroform  was  of  a  syrupy 
consistence.  These  residues  were  treated  with  distilled  water, 
which  almost  entirely  dissolved  them.  The  following  reagents 
were  applied  to  the  filtered  solutions — both  of  which  gave  the  same 
reactions  : 
Blue  litmus  paper,  reddened. 
Silver  nitrate,  reduced. 
Gold  chloride,  reduced. 
Barium  chloride,  no  change. 
Lead  acetate,  precipitate. 
Lime  water,  no  change. 
Ferric  chloride,  dark  blue  color. 
Ferric  alum,  dark  blue  color. 
Potassium  bichromate,  dark  red  color. 
Fehling's  solution,  before  treatment  with  acid,  slight  reduction. 
Fehling's  solution,  after  treatment  with  acid,  no  increase. 
These  results  indicate  the  presence  of  an  organic  acid. 
The  material  removed  from  the  alkaline  aqueous  liquid  by  agita- 
ting it  with  ether  was  crystalline  and  had  a  narcotic  odor.  The 
corresponding  substance  obtained  from  this  solution  by  the  use  of 
chloroform  was  a  dark-colored  thick  liquid,  possessing  a  narcotic 
odor.  These  products  were  then  dissolved  in  acidified  water.  To 
separate  portions  of  each  of  these  solutions,  Mayer's  reagent,  tannic 
acid,  potassium  tri-iodide,  gold  chloride,  picric  acid,v  platinic  chlor- 
ide and  phosphotungstic  acid  were  added,  all  of  which  indicated 
the  presence  of  an  alkaloid.  These  experiments  were  repeated 
with  the  same  results  on  2  kilos  of  the  powdered  plant.  A  small 
portion  of  the  residue,  instead  of  being  tested  with  the  foregoing 
reagents,  was  added  to  some  soda-lime  contained  in  a  test  tube.' 
