856 
The  Theory  of  Percolation. 
I  Am.  Jour.  Pharm. 
I     December,  1920. 
Table  I. 
Percolation  of  7680  Grains  of  Cimicifuga. 
Fluidounce  of 
Gms.of 
Fluidounce  of 
Gms.  of 
Percolate. 
Dry  Extract. 
Percolate. 
Dry  Extract 
I 
3-33 
13 
0.  77 
2 
2.80 
14 
0.68 
3 
2.38 
15 
0,58 
4 
2.19 
16 
0.35 
5 
I  94 
17 
0.36 
6 
I  .71 
18 
0.32 
7 
1-43 
19 
0.23 
8 
1-34 
20 
0 . 22 
9 
I  .  12 
2 1 
0 . 22 
10 
0.84 
22 
0,21 
1 1 
0.79 
23 
0 . 22 
12 
0.73 
24 
0.21 
These  results  have  been  plotted  as  Chart  B  and  show  in  a  striking 
manner  the  uniformity  of  the  decreasing  extract  content  in  the 
percolate.  Many  similar  results  from  the  work  of  Squibb,  Diehl, 
Robbins  and  others  confirm  the  fact  of  regularity  in  extraction. 
When,  however,  we  subject  the  experimental  results  to  mathe- 
matical treatment  we  discover  no  expression  which  applies  to  the 
conditions  of  every  fraction.  Instead,  in  the  case  of  every  formula 
which  one  would  expect  to  generalize  the  phenomena,  we  find  vari- 
ation where  we  should  obtain  constancy,  and,  furthermore,  the 
variation  appears  to  proceed  in  a  regular  fashion  from  one  fraction 
of  the  percolate  to  the  next  equal  fraction,  either  increasing  or  de- 
creasing the  "constant,"  which  indicates  that  a  constantly  varying 
factor  or  set  of  factors  is  missing  from  our  data. 
What  these  missing  factors  are  can  only  be  indicated  here. 
I  hope  to  consider  this  phase  of  the  subject  more  in  detail  at  some 
future  time  when  I  may  present  new  data.  The  first  factor  which 
may  vary  is  the  chemical  composition  of  the  extract  contained  in 
each  fraction.  From  the  evidence  presented  under  the  discussion 
of  the  composition  of  the  precolate  we  should  expect  the  first  frac- 
tions of  the  percolate  to  differ  qualitatively  from  those  fractions  ob- 
tained toward  the  end  of  the  process.  This  question  was  investi- 
gated by  Squibb  on  cinchona^  and  on  podophyllum.'  He  states: 
"The  extract  or  soluble  matter  yielded  to  the  menstruum  is  not 
^  This  Journal,  Vol.  39,  402,  (1867). 
■  2  This  Journal,  Vol.  40,   i,  (1868). 
