198  Arctostaphylos  Glauca,  Lindley.  {^g^Sf"" 
posited,  soluble  in  water,  alcohol  and  ether ;  sparingly  so  in  chloro- 
form, petroleum  benzin  and  bisulphide  of  carbon.  Each  of  these 
menstrua  was  tried  as  a  medium  for  obtaining  crystals,  but  without  a 
favorable  result. 
The  precipitate  with  neutral  acetate  of  lead  was  diffused  in  water, 
decomposed  by  hydrosulphuric  acid,  the  sulphide  of  lead  removed. 
The  filtrate  gave  no  evidence  of  organic  matter  when  evaporated  on 
platinum  foil.  Barium  chloride  gave  no  precipitate,  but  on  the  addi- 
tion of  ammonia  a  yellowish  precipitate  was  thrown  down,  redissolved 
in  an  excess.  Nitrate  of  silver  gave  a  yellowish  precipitate,  dissolved 
on  the  addition  of  ammonia.  No  precipitate  was  produced  with  ferric 
chloride. 
The  leaves  yielded  42  per  cent,  of  soluble  matter  to  boiling  water. 
The  amount  of  tannin  was  ascertained  by  volumetric  analysis  with 
solution  of  gelatin  to  be  94b  per  cent.  After  incinerating  the  air  dry 
leaves,  6  per  cent,  of  ashes  was  left  as  a  residue,  containing  potas- 
sium, calcium,  magnesium  and  iron. 
A  second  portion  of  leaves  was  reduced  to  a  fine  powder,  displaced 
with  alcohol,  the  percolate  evaporated  to  an  extract,  this  treated  with 
hot  water,  the  residue  separated  by  a  filter,  the  solution  precipitated 
with  neutral  acetate  of  lead,  filtered  and  washed.  Through  the  fil- 
trate and  washings  hydrosulphuric  acid  was  passed,  the  sulphide  of 
lead  was  removed  by  a  filter,  and  the  liquid  evaporated  to  the  consist- 
ence of  an  extract.  This  was  digested  with  ether,  and  the  filtrate 
allowed  to  evaporate  spontaneously. 
After  standing  several  days,  a  crystalline  mass  was  deposited,  hav- 
ing the  appearance  of  white  crystals  diffused  in  a  brownish-yellow 
coloring  matter.  This  was  proved  to  be  very  soluble  in  water,  alco- 
hol and  ether ;  slightly  so  in  chloroform,  benzin  and  bisulphide  of 
carbon. 
A  portion  was  treated  with  each  of  these  solvents,  and  shaken  with 
animal  charcoal,  filtered  and  set  aside  to  crystallize. 
When  first  filtered  the  solution  was  clear,  but  after  standing  a  few 
hours  re-assumed  the  original  color  (brownish-yellow). 
An  aqueous  solution  of  the  residue  left,  after  treating  with  ether, 
gave  the  characteristic  precipitate  of  suboxide  of  copper  with  Trora- 
mer's  test  for  grape  sugar. 
A  minute  quantity  of  the  crystalline  product  was  dissolved  in  water, 
and  the  solution  rendered  alkaline  by  ammonia,  when  phosphomolyb- 
