500  Astrmgencies  of  Red  and  Pale  Rose.    \  AmjJiy,ris^i.arm" 
stituents  and  of  a  crystalline  substance  having  the  same  appearance 
and  solubilities  as  the  one  being  sought. 
Continuing  the  experiment  the  ether  exhausted  residue  (un- 
changed in  appearance)  was  freed  of  this  solvent,  and  mixed  with 
water  into  a  mush,  which  was  exposed  to  induce  a  growth  of  mildew. 
In  the  course  of  two  weeks,  the  surface  was  covered  with  a  thick 
layer  of  mycelium.  This  covering  was  removed  with  as  little  of  the 
Red  Rose  as  possible,  and  extracted  with  ether ;  which  removed  but 
a  trifle  of  the  crystalline  substance. 
The  mush  was  then  strained  to  separate  the  aqueous  portion, 
and  this  clarified  by  further  straining.  This  liquid  was  a  deep  wine- 
red  color,  strongly  acid  to  litmus,  astringent,  bitter,  and  decidedly 
musty,  but  still  strongly  suggestive  of  rose.  Ether  shaken  with  this 
fluid  removed  the  crystalline  principle,  thus  confirming  its  forma- 
tion under  the  circumstances  arranged  for.  From  this  portion  of  the 
mush,  the  yield  was  several  times  what  ether  extracted  directly  from 
the  Red  Rose. 
It  is  presumable  that  this  crystalline  substance  is  derived  from 
some  water-soluble  constituent  of  the  rose,  whether  the  "tannin"  or 
not.  To  examine  this  subject,  the  solid  portion  of  the  mush  was 
washed  with  cold  water  while  ever  color  was  removable ;  then  mixed 
again  with  water  and  the  resulting  mush  exposed  to  induce  mildew  as 
before.  The  development  of  mildew  was  very  slow  and  sparce  com- 
pared to  its  appearance  and  amount  in  the  previous  experiment.  The 
watery  portion  was  found  to  contain  none  of  the  crystalline  prin- 
ciple. 
The  same  experiments  were  carried  out  on  Pale  Rose  with  the 
results  that  a  small  quantity  of  a  crystalline  substance,  apparently  the 
same  as  that  from  Red  Rose  was  directly  extracted  by  ether ;  while  a 
less  amount  was  obtained  in  the  experiment  with  mildew. 
These  experiments  shall  be  repeated  to  determine  whether  the 
crystals  will  develop  in  the  mush  without  the  appearance  of  mildew. 
It  is  within  the  bounds  of  probability  that  this  crystalline  sub- 
stance is  related  to  the  astringent  principle. 
The  solubility  of  this  crystalline  principle  in  chloroform  distin- 
guishes it  completely  from  the  astringent  principle. 
The  burr-like  aggregates  melt  when  heated  and  sublime  in  crys- 
talline form,  tending  to  re-assume  this  peculiar  manner  of  association. 
