AmMay''Xh8arm'}      Fluid  Extracts  by  Reper eolation.  249 
percolator,  the  exhausted  residue  should  be  so  removed  as  not  to 
disturb  the  disks  of  paper  and  blanket  at  the  bottom. 
The  principles  of  this  process  once  well  understood,  modifications 
of  apparatus  will  occur  to  many.  The  simplest  of  those  that  has  been 
tried  on  a  scale  larger  than  the  one  above  shown  with  lamp  chimneys, 
is  to  dispense  with  well  tube  and  syphon,  and  replace  them  with  a 
piece  of  rubber  tubing  of  small  bore.  One  end  of  this  is  placed 
between  the  two  disks  of  flannel  near  the  centre  of  the  percolator, 
and  then  the  tubing  is  led  up  through  any  part  of  the  packed  substance, 
- — say  near,  but  not  against  the  side  of  the  percolator,  as  this  would 
leave  channels  for  liquid, — and  then  out  over  the  edge  of  the  perco- 
lator. Then  a  small  bent  portion  of  glass  tubing  is  slipped  into  the 
end  of  the  rubber  so  as  to  represent  the  end  of  the  glass  syphon. 
This  end  can  then,  by  the  flexibility  and  length  of  the  rubber  tube,  be 
kept  at  any  desired  position.  This  however  does  not  answer  as  well  in 
practice  as  the  well  and  syphon,  nor  does  any  other  yet  tried,  including 
the  simplification  adopted  with  the  lamp  chimneys, — answer  as  well, 
when  judged  by  the  results  obtained. 
The  writer  made  a  conditional  promise  at  the  request  of  the  Com- 
mittee that  he  would  give  a  table  showing  his  own  practice  with  fluid 
extracts  in  regard  to  the  menstruum  now  used  for  each, — the  weight 
of  a  pint  of  the  menstruum  and  the  weight  of  a  pint  of  the  finished 
fluid  extract  in  each  case,  as  bearing  upon  the  proposed  new  relation 
of  making  them  weight  for  weight  instead  of  minim  for  grain,  and  the 
work  for  constructing  such  a  table  has  been  done.  But  this  paper  has 
grown  to  such  an  unreasonable  length  that  it  will  hardly  be  read,  and 
the  calculations  and  construction  of  the  table  would  require  so  much 
additional  time  that  the  writer  must  beg  the  Committee  to  excuse  him 
for  not  presenting  it. 
The  writer  was  rather  opposed  to  the  new  relation  of  weight  for 
weight  when  this  point  was  discussed  by  the  Committee,  but  now 
considers  it  practicable  if  the  labor  be  given  to  make  it  fairly  accurate  ; 
and  believes  that  it  might  be  made  far  more  accurate  than  the  present 
relation  of  minim  for  grain,  this  latter  having  proved  to  be  rather  an 
ideal  than  a  practically  true  relation. 
Brooklyn^  April  16th,  1878. 
9 
