226  AN  ALKALOID  IN  LYCIUM  BARBARUM. 
be  rich  in  the  alkaloid.    The  decoctions  obtained  by  boiling  the 
parts  properly  comminuted  in  repeated  portions  of  water,  and 
which  were  pretty  dark-colored,  were  precipitated  by  subacetate 
of  lead.     The  copious  lemon-colored  precipitate  was  separated 
from  the  liquor  by  decantation,  and  the  latter,  after  being  freed 
from  excess  of  lead  by  means  of  sulphuric  acid  and  again  neu- 
tralized with  soda,  was  evaporated  to  one-third  of  its  original 
bulk.    It  was  then  again  strongly  acidulated  with  sulphuric 
acid  and  completely  precipitated  by  phosphomolybdate  of  soda. 
As  we  had  to  use  large  quantities  of  this  reagent,  we  prepared  it, 
not  according  to  the  method  directed  by  Sonnenschein,  but  by 
simply  dissolving  30  equivalents  of  molybdic  acid  in  a  solution  of 
the  corresponding  quantity  of  caustic  soda,  adding  afterwards  an 
aqueous  solution  of  one  equivalent  of  phosphate  of  soda.  The 
pale  yellow  colored,  flocculent  precipitate  deposits  very  readily, 
and  can  then  be  washed  with  water  acidulated  with  sulphuric 
acid  without  undergoing  any  perceptible  decomposition. 
The  high  price  of  molybdic  acid  induced  us  to  employ  in  its 
stead  the  tungstate  of  soda  now  to  be  had  in  commerce  at  one- 
fifth  the  cost  of  the  other,  and  the  same  object  is  attained  in 
fact  by  employing  a  mixture  of  30  equivalents  of  tungstate  and 
one  of  phosphate  of  soda.  But,  taking  into  account  that  a 
complete  precipitation  by  phosphotungstic  acid  is  possible  only 
in  presence  of  a  large  excess  of  sulphuric  acid,  that  the  precipi- 
tate is  much  more  difficult  to  wash  than  the  phosphomolybdate 
of  the  alkaloid,  and  that  in  view  of  the  difference  in  the 
equivalents,  that  in  expense  is  after  all  only  as  1  :  2 :  it  is 
probable  that  the  molybdic  compound  will  be  maintained  in  use, 
until  tungstate  of  soda,  as  is  quite  likely,  shall  have  become 
less  expensive. 
The  precipitates  formed  by  either  of  these  reagents  were, 
while  still  moist,  mixed  with  carbonate  of  baryta,  the  pasty  mass 
on  the  water  bath  with  constant  stirring  reduced  to  dryness,  and 
then  exhausted  with  boiling  alcohol.  The  alcoholic  solutions  on 
evaporation  left  a  somewhat  colored  syrupy  residue,  without 
signs  of  crystallization,  the  solution  of  which  in  dilute  hydro^ 
•chloric  acid,  after  filtering  off  some  resinous  matter,  yielded  on 
evaporation  very  fine  crystals.  When  pressed  between  bibulous 
paper  and  repeatedly  recrystallized  from  hot  alcohol,  these  form 
