7  8  Alkaloids  of  the  Aconite  Roots.       { Am'fcl^m 
these  specimens,  however,  did  not  crystallize  in  the  well-defined 
rhombohedrons  exhibited  by  the  other  substance.  All  these  specimens 
appeared  to  be  eminently  toxic  :  besides  "  pseudaconitia,"  amorphous 
or  difficultly  crystallizable  alkaloids  were  obtained  from  the  alcoholic 
extract ;  but  as  yet  these  have  not  been  fully  examined. 
Roots  of  Aconitum  Napellus^  when  treated  in  the  same  way,  yielded  a 
mixture  of  at  least  two  alkaloids,  forming  readily  crystallizable  salts, 
besides  other  amorphous  bases.  Of  these  two  bases,  one  partially 
separated  as  a  hydrochloride  at  a  certain  stage  during  the  working  up 
of  the  etherial  solution  ;  this  hydrochloride,  when  purified  by  succes- 
sive crystallizations,  yielded  numbers  agreeing  best  with  the  formula 
C31H45NO10.HCl.iiH2O,  the  formula  of  the  free  base  being  C31H45NO10, 
and  that  of  the  gold  salt  C31H45NO10. HC1. AuCl3.  The  platinum  salt 
of  this  base  is  also  somewhat  readily  soluble.  Although  the  salts  (e.  g., 
the  nitrate  and  hydrochloride)  of  this  base  crystallize  with  great  facility, 
it  was  not  found  practicable  to  obtain  the  free  base  in  a  crystalline 
form  ;  from  all  solvents  tried,  it  separated  only  as  a  varnish.  Com- 
paratively speaking,  this  base  is  inert  ;  its  salts  do  not  produce  the 
tongue-prickling  characteristics  of  aconite  roots,  but  have  a  pure  bitter 
taste  ;  Mr.  Groves  found  that  half  grain  doses  taken  by  himself  pro- 
duced no  noticeable  effects. 
The  liquor  from  which  this  hydrochloride  separated  contained, 
besides  a  large  quantity  of  this  non  crystallizable  base,  a  second 
alkaloid  separating  from  the  etherial  solution  of  the  mixed  bases  by 
spontaneous  evaporation  in  anhydrous  crystals  ;  after  several  crystalli- 
zations from  ether,  numbers  were  obtained  leading  to  the  formula 
C33H45NOn,  the  air-dry  hydrochloride  being  C33H45NOn,HCl,3H20. 
This  substance  agrees  in  physiological  and  physical  characters  with  the 
"  aconitia  "  of  Duquesnel  and  others  ;  Duquesnel,  however,  attributed 
to  crystallized  "aconitia"  from  A.  Napellus  the  formula  C27H40NOl0 
("  Compt.  rend.,"  Ixxiii,  207),  which  requires  very  different  numbers 
from  those  found  by  the  authors  ;  his  mode  of  preparation,  however, 
was  somewhat  different  from  that  adopted  by  Mr.  Groves,  alcoholic 
tartaric  acid  being  employed  instead  of  alcoholic  hydrochloric  acid,  and 
the  extract  being  evaporated  at  a  temperature  not  exceeding  6o°. 
Putting  the  alkaloid  through  the  mercuric  iodide  process  seems  to 
lower  its  molecular  weight. 
It  seems  highly  probable,  on  comparing  the  above  results  with  those 
of  others  who  have  worked  on  the  subject,  that  the  aconite  roots  con 
