434  Scammony  and  Resin  Scammony.  {^?pimbre'rPi9iodi 
numbers.  As  stated  before,  the  moisture  content  of  these  resins 
was  not  reduced  by  the  method  of  drying  adopted  by  us.  The  re- 
sults agree  perfectly  with  those  found  by  Cowie  and  Taylor  and  no 
further  comment  is  necessary,  except  to  call  attention  to  the  high 
percentage  of  moisture  in  the  commercial  samples  I  and  VII. 
The  percentage  of  ash  is  rather  constant,  and  in  all  cases  except 
I  is  well  below  the  limit  of  i  per  cent,  allowed  by  the  U.S. P.  for 
resin  scammony.  The  percentage  of  ash  found  in  I  is  also  below 
the  limit  of  3  per  cent,  allowed  by  the  U.S. P.  for  virgin  scammony; 
this  high  percentage  of  ash  confirms  our  belief  that  the  resin  had 
been  only  superficially  purified  by  straining  from  gross  impurities. 
Nothing  definite  can  be  concluded  from  the  acid  numbers,  as  has 
already  been  stated  by  Taylor  and  Cowie.  The  saponification  num- 
bers, on  the  other  hand,  fall  into  two  well-defined  groups.  I,  II, 
III  have  saponification  numbers  which  are  considerably  over  200. 
II,  which  may  be  taken  as  a  representative  sample  of  purified  virgin 
scammony,  has  the  value  236.6  which  is  in  close  agreement  with  the 
average  of  about  238  found  by  Taylor,  Cowie,  and  others.  The 
saponification  value  of  the  other  five  samples  range  within  narrow 
limits  around  the  number  177,  which  is  lower  than  the  average 
saponification  value  obtained  by  Taylor  and. Cowie. 
Table  II  shows  the  solubility  in  various  solvents.  The  most  im- 
portant results  are  the  solubility  in  ether.  Cowie  and  Guigues  have, 
as  already  stated,  directed  attention  to  the  necessity  of  using  anhy- 
drous ether  in  determining  the  solubility,  and  their  results  are  fully 
confirmed  by  the  present  work. 
Sample  I,  the  most  impure  specimen  of  all,  was  soluble  to  the 
extent  of  71.8  per  cent,  in  dry  ether,  but  to  the  extent  of  85  per  cent, 
in  commercial  ether,  thus  passing  the  requirements. 
Sample  II,  the  purified  resin  prepared  from  I,  was  entirely 
soluble  both  in  anhydrous  and  commercial  ether.  Ill  was  soluble 
to  the  extent  of  96  per  cent,  in  commercial  ether,  and  by  stretching 
the  words  "  almost  completely  "  might  have  passed  muster  as  the 
U.S. P.  article,  although  it  was  prepared  by  percolating  scammony 
root.  The  other  five  specimens  vary  between  80  and  90 
per  cent,  soluble  in  hydrous  ether,  and  were  all  nearly  90  per  cent, 
soluble  in  anhydrous  ether.  In  this  connection  we  must  call  atten- 
tion to  the  phenomenon  always  noticed  when  dissolving  Mexican 
scammony  in  ether.  Using  U.S. P.  ether,  containing  varying 
amounts  of  water  and  alcohol,  a  part  of  the  resin  invariably  separates 
