124   PROPOSED  ECONOMY  OF  ALCOHOL  IN  PERCOLATION,  ETC. 
saja  bark  may  contain  on  an  average  four  per  cent,  of  its  weight 
of  alkaloids  in  their  natural  combinations,  and  this  small  pro- 
portion of  the  whole  is  alone  medicinally  effective,  so  that  all 
the  remainder  should  be  excluded  from  medicinal  use,  if  that 
was  practicable.  But  calisaya  bark,  by  the  officinal  process  of 
percolation,  yields  twenty-two  per  cent,  of  its  weight  of  dry 
extract.  It  follows,  then,  that  at  least  four-fifths,  or  eighty  per 
cent.,  of  this  dry  extract  is  inert  and  useless,  and  need  not  be 
extracted  if  the  remaining  fifth,  or  twenty  per  Cent.,  could  be 
extracted  without  it. 
Now,  if  we  admit,  in  order  to  secure  perfectly  safe  conclusions, 
that  in  the  percolation  the  exhaustion  of  the  active  medicinal 
portion  of  the  extract  is  not  more  easy  or  more  rapid  than  that 
of  the  inert  portion,  but  that  it  diminishes  in  the  same  ratio, 
then  it  is  only  necessary  to  subtract  eighty  per  cent,  from  the 
weight  of  extract  yielded  from  each  successive  portion  of  per- 
colate to  obtain  a  safe  and  useful  index  of  the  true  medicinal 
value  of  every  stage  of  the  process  of  percolation.  And  this 
once  obtained,  shows,  with  something  like  mathematical  accu- 
racy, where  the  percolation  may  be  stopped  to  obtain  the  best 
and  most  economical  practical  results. 
The  following  table  (Table  II.)  presents  the  results  of  three  ex- 
periments with  the  old  bark  and  one  with  the  new.  The  first  three 
vary  only  in  the  proportion  of  menstruum  used  to  moisten  the  pow- 
der before  packing.  The  middle  proportion,  viz.,  ten  fluidounces, 
is  the  officinal  one  ;  but  the  last  proportion,  namely,  eight  fluid- 
ounces,  was  found  to  give  the  best  results,  and  was  therefore 
adopted  in  the  fourth  experiment  with  the  new  bark.  This  ex- 
periment with  the  new  bark  was  repeated  three  times,  and  of 
these,  the  lowest  or  least  favorable  one  was  chosen  for  the  table. 
The  first  column  of  the  table  indicates  the  portion  of  the  perco- 
late represented  by  the  quantity  of  dry  solid  extract  upon  the 
same  line  in  all  the  succeeding  columns.  Then,  each  of  the 
four  experiments  has  a  group  of  three  columns.  The  first  of 
these  three  gives  the  number  of  grains  of  dry  solid  extract  con- 
tained in  the  measure  of  percolate  opposite  to  which  it  is  found. 
The  second  column  of  the  group  gives  the  percentage  amount  of 
this  dry  solid  extract  as  calculated  from  the  total  extract  yielded 
