72        The  Coloring  Principle  of  Eraser  a  Walteri,  {"^""f^b^is^so"" 
filtered,  then  submitted  to  a  temperature  ranging  from  o°F.  to  — lO^F. 
for  the  period  of  thirty-six  hours,  at  the  end  of  which  the  sides  and 
bottom  of  the  vessel  were  studded  with  a  crystalh'ne  deposit  consisting 
of  two  distinct  classes  of  crystals,  one  class  hard  and  transparent,  the 
other  needle-like  and  yellow. 
The  transparent  were  sweetish,  contaminated  with  bitterness,  soluble 
in  water,  slightly  in  alcohol,  insoluble  in  ether.  After  trituration  with 
repeated  portions  of  alcohol  (to  remove  grape  sugar  and  adhering  bitter 
principles)  they  were  dissolved  in  water.  The  solution  was  purely  sweet, 
refused  to  be  affected  by  Fehling's  solution  and  gave  no  reaction  with 
tests  for  the  alkaloids.  Upon  exposing  the  solution  in  a  watch  crystal 
to  a  current  of  air,  I25°F.,  crystals  of  the  shape  of  cane  sugar  were 
deposited.  These  were  charred  by  concentrated  sulphuric  acid.  This 
substance  was  cane  sugar.    (See  specimen.) 
1  he  second  class  of  crystals  were  in  beautiful  lemon-yellow  tufts 
radiating  in  silky  needles  from  a  common  centre.  They  were  bitter 
when  tirst  separated  from  the  mother  solution  and  acid  to  litmus  paper, 
both  of  these  characteristics  depending  upon  the  presence  ot  adhering 
impurities.  They  were  washed  repeatedly  in  cold  alcohol,  in  which 
they  dissolved  but  slightly,  then  with  warm  water  until  free  from  tugar. 
As  thus  obtained  (see  specimen  A)^  they  were  lemon-yellow,  odorless, 
tasteless.  They  were  insoluble  in  cold  water,  but  tinged  boiling  water 
slightly  straw  color.  They  dissolved  slightly  in  cold  alcohol,  freely  in 
boiling  alcohol,  very  freely  in  cold  ether,  cold  chloroform  and  cold 
carbon  disulphide,  more  being  dissolved  by  each  menstruum  which 
followed  that  preceding.  The  crystals  were  neutral  to  moistened  lit- 
mus paper,  red  or  blue,  and  all  of  the  solutions  failed  to  affect  either 
colored  paper.  When  the  crystals  were  boiled  with  water,  the  addition 
of  ferrous  sulphate  and  ferrous  chloride  failed  in  both  cases  to  produce 
a  precipitate  or  alter  the  color  of  either  crystal  or  solution  towards  blue 
or  green. 
With  nitric  acid  the  crystals  dissolve,  forming  a  deep-red  solution 
without  evolution  of  nitric  oxide.  With  sulphuric  acid  they  form  a  deep 
orange-red  solution.  With  hydrochloric  acid  in  the  cold,  no  apparent 
change  ;  upon  boiling,  the  acid  becomes  straw-colored,  while  the  crys- 
tals seemingly  remain  unaltered.  They  refused  to  unite  with  either  of 
the  aforementioned  acids  when  dilute. 
Upon   boiling   the  purified   crystals  some  time   with  dilute  sul- 
