On  the  rower  of  Soils  to  absorb  Manure.  339 
there  is  no  security  that  in  percolating  it,  when  used  as  a filter, 
the  liquid  shall  come  in  contact  with  every  particle.  The  ten- 
dency to  form  channels  by  which  the  subsequent  portions  of  the 
liquid  would  pass,  leaving  other  parts  of  the  filter  comparatively 
unacted  on,  is  greatly  opposed  to  the  chance  of  a satisfactory  result 
being  obtained  by  this  method ; and  in  addition  to  this  primary 
difficulty,  there  are  others  which  would  occur  to  any  person  making 
such  experiments,  and  which  need  not  here  be  further  discussed. 
It  is  obvious  that  the  power  of  a soil  to  absorb  and  solidify  various 
alkaline  substances  is  independent  of  the  accidental  circumstances 
of  filtration,  and  would  be  exhibited  whether  the  soil  or  the  solution 
were  used  in  excess.  If  then  we  have  a solution,  say  of  ammonia 
of  known  strength,  and  that  we  mix  with  any  given  quantity  of  it 
a known  weight  of  the  soil  to  be  examined,  the  quantity  of  am- 
monia in  the  liquid  will  be  reduced  by  so  much  as  the  soil  has 
absorbed.  By  examining  the  clear  liquid  after  the  soil  has  been 
allowed  to  fall  to  the  bottom,  we  shall  at  once  have  the  means  of 
calculating  what  quantity  of  ammonia  it  has  lost,  and  consequently 
what  quantity  the  soil  has  taken  up.  In  experimenting  upon 
ammonia,  which  it  is  well  known  is  very  volatile,  it  was  necessary 
to  guard  against  loss  during  the  time  occupied  in  the  experi- 
ment. After  several  attempts,  a simple  apparatus  was  contrived, 
by  which  the  solution  of  ammonia  and  the  soil  could  be  left  in 
contact  for  any  length  of  time,  and  the  clear  liquid  filtered  off 
without  the  slightest  exposure  to  loss.  The  ammoniacal  solu- 
tion, as  it  fell  from  the  apparatus,  was  received  in  a small  bottle 
containing  a weighed  quantity  of  hydrochloric  acid,  and  the 
quantity  to  be  employed  in  the  subsequent  operations  could  then 
be  determined  with  perfect  safety. 
It  will  be  thus  seen  that  the  plan  adopted  has  been  that  of 
determining  the  quantity  of  base  absorbed  by  that  which  the  so- 
lution has  lost.  It  does  not  appear  that  this  method  is  open  to 
much  objection,  and  it  is  far  preferable  to  the  only  other  that 
could  be  resorted  to — namely,  the  examination  of  the  soil  itself 
after  subjecting  it  to  the  action  of  the  different  salts.  Indeed, 
although  this  latter  plan  may  seem  at  first  sight  the  most  simple 
and  direct,  it  is  upon  closer  inspection  found  to  be  quite  inad- 
missible in  most  cases.  Let  us  suppose  that  it  is  wished  to  ascer- 
tain the  absorbent  power  of  any  soil  for  ammonia.  After  placing 
it  in  contact  with  a solution  of  this  alkali,  we  require  to  dry  it,  in 
order,  by  burning  it,  to  drive  off  the  ammonia.  What  security  is 
there  that  the  act  of  drying  will  not  drive  off  the  ammonia  ? Then 
again,  having  surmounted  this  difficulty,  we  meet  with  another. 
It  is  to  be  expected  that  as  this  power  exists  in  soils,  it  will  mani- 
fest itself  naturally,  and  every  soil  will  more  or  less  contain  am- 
z 2 
