794  The  Theory  of  Percolation.         { XviZer!"''i92o; 
The  rates  at  which  the  various  constituents  of  a  drug  go  into  solu- 
tion appear  to  differ  materially;  certain  substances,  especially  those 
which  are  crystalline  and  of  low  molecular  weight,  diffuse  more  rapidly 
than  cumbrous,  amorphous  compounds  and  so  are  found  in  larger 
quantity  in  the  first  portions  of  percolate  than  in  later  fractions. 
Our  investigations  of  this  most  important  subject  have  merely  dipped 
below  the  surface.  Squibb's  work,  quoted  in  detail  below,  has  estab- 
lished the  fact.  In  a  drug  like  prunus  serotina  from  which  Power 
and  Moore^  isolated  no  less  than  fourteen  constituents  varying  in 
character  from  a  water-soluble  glucoside  to  resin  and  phytosterol, 
the  relative  rates  of  extraction  with  the  ordinary  pharmacopoeial 
menstruum  must  be  very  diverse,  and  to  this  is  added  the  possibility 
of  the  complete  hydrolysis  of  the  glucoside  yielding  hydrocyanic 
acid  which,  by  altering  the  hydrogen-ion  concentration  of  the  solution, 
might  cause  precipitation  of  the  resin. 
Moreover,  as  the  menstruum  proceeds  downward  and  the  dis- 
solving of  the  extractive  decreases  in  extent,  the  rate  of  solution  for 
any  individual  substance  must  be  diminished,  but  there  is  no  reason 
to  expect  that  the  rates  of  solution  of  all  the  soluble  constituents  are 
affected  in  the  same  proportion.  Consequently,  this  is  another  condi- 
tion which  may  vary  the  composition  of  the  precolate. 
In  addition,  the  mutual  effects  of  dissolved  substances  upon  each 
other  and  upon  soluble,  but  undissolved,  material  affect  the  composi- 
tion of  the  precolate.  Certain  substances  are  much  less  soluble  in 
solutions  of  some  other  compounds  than  they  are  in  the  pure  solvent. 
Glucosides  are  less  soluble  in  slightly  acid  liquids  than  they  are  in 
neutral  or  alkaline  media.  Terpene  oils  are  quite  insoluble  in  liquids 
which  contain  inorganic  salts.  Under  such  conditions  the  less  soluble 
constituent  would  be  obliged  to  remain  in  the  marc  until  the  greater 
part  of  the  precipitant  is  extracted  before  it  may  be  dissolved.  In  a 
case  of  this  kind  we  should  have  qualitatively  different  solutions  in 
the  upper  and  lower  regions  of  the  percolator. 
Contrariwise,  many  constituents  of  drugs  are  extractible  by  sol- 
vents in  which  they  are  really  insoluble  because  of  the  presence 
in  solution  of  other  substances  which  modify  their  solubility.  Sugars 
are  readily  extracted  by  alcohol  although  they  are,  as  a  class,  but 
little  soluble  in  that  liquid.  Glucosides  which  are  insoluble  in  water 
may,  nevertheless,  often  be  extracted  by  that  solvent.  Gitalin,^ 
1  /.  Chem.  Soc.  95,  243-261,  (1909). 
2  Meyer,  Arch.  exp.  Path.  u.  Pharm.  81,  261,  288,  (1917). 
