EXPERIMENTS  WITH  PAPER  FILTERS. 
201 
opening  water-ways  between  the  glass  and  the  paper  ;  the  inter- 
position of  straws,  glass  rods  and  splinters  of  wood  between 
filter  and  funnel,  as  well  as  fluted  funnels  and  plaited  filters  are 
all  devices  looking  to  this  end. 
The  advantages  of  the  plaited  filter  are  so  great  that  some 
chemists  prefer  to  use  it,  even  in  quantitative  analysis,  instead 
of  the  common  form,  in  spite  of  its  greater  liability  to  break,  and 
the  difficulty  of  washing  the  precipitate. 
In  the  laboratory  of  Profs.  Lawrence  Smith  &  Ordway, 
among  the  most  accurate  of  American  analysts,  plaited  filters 
are  said  to  be  employed  to  the  almost  total  exclusion  of  the 
plain  form. 
Another  excellent  method  of  increasing  the  speed  of  filtration, 
first  suggested  in  this  country  by  the  German  chemist,  Fleit- 
mann,  consists  in  placing  one  plain  filter  within  another  of 
coarser  fibre ;  for  instance,  a  fine  plain  filter  of  Swedish  paper 
may  be  placed  within  another  plain  filter  of  coarse  German  pa- 
per, supported,  as  usual,  on  a  funnel. 
In  experimenting  upon  these  various  forms  of  filters,  it  occurred 
to  me  to  fold  the  plain  qualitative  filter  in  two  operations  instead 
of  one.  In  place  of  folding  the  filter  doubled  upon  itself  down 
the  middle  in  the  usual  way,  I  proposed  to  turn  down  on  each 
side  of  the  paper  a  fold  equal  to  one-quarter  of  the  semi-circle, 
and  then  to  fold  the  sectors  of  45°  arc  thus  formed  back  upon 
themselves. 
The  filter  is  then  opened  without  disturbing  the  folded  por- 
tions, and  placed  upon  the  funnel.  In  this  form  the  triple  side 
of  the  plain  filter  is  broken  up  and  the  folded  portions  keep  open 
passages,  instead  of  hindering  filtration. 
This  filter,  as  tried  against  the  plain  form,  gave,  1st,  133  :  100. 
2d,  111  +  :  100.    3d,  205  +  :  100. 
Two  plain  filters  ran  equally  in  several  trials ;  each  was 
changed  into  the  other's  funnel,  and  No.  1  ran  33  per  cent,  less 
than  No.  2.  No.  1  was  dried  and  folded  into  my  form  ;  remain- 
ing in  the  same  funnel,  it  ran  32  per  cent,  faster  than  the  other. 
Both  filters  were  then  opened,  and  showed  no  tear  or  weakness 
when  held  against  the  light. 
As  these  filters  gave  different  results  in  different  funnels,  I 
thought  I  would  ascertain  the  cause.    The  water  seemed  to  be 
