322 
On  the  Potver  of  Soils  to  absorb  Manure. 
Experiment  G. — The  same  soil  passed  through  a 40-hole  sieve,  but  the 
fine  portions  not  removed  as  in  the  previous  experiments.  A much 
weaker  solution  of  salt,  employed.  The  very  first  drops  of  liquid  gave,  by 
nitrate  of  silver,  a copious  precipitate. 
From  these  experiments  it  is  obvious  that  whatever  may  be 
the  nature  of  the  action,  it  is  a bona  fide  absorption  of  the 
ammonia  of  the  soil.  The  result  would  seem  to  prove  incident- 
ally that  common  salt  is  not  subject  to  any  absorption  by  filtration 
through  a soil  ; but  experiments  to  be  presently  detailed,  serve 
materially  to  modify  such  an  opinion. 
The  following  experiment  on  dried  soil  is  confirmatory  of  those 
that  were  made  with  common  salt ; it  also  proves  that  the  absorp- 
tive power  is  not  destroyed  by  drying : — 
Experiment  7. — A quantity  of  the  same  soil  was  dried  on  a sand-bath 
for  many  hours  at  a heat  varying  between  130°  and  160°  F.  A solution 
of  ammonia  (of  the  same  strength  as  in  Experiment  No.  1)  was  poured  upon 
a column  of  14  inches,  weighing  2800  grains.  Owing  to  the  finely  divided 
dusty  state  into  which  the  drying  had  brought  the  soil,  the  solution  took 
three  hours  to  percolate  ; when  it  came  through  it  was  entirely  free  from 
ammonia,  and  so  continued  until  fully  1 ounce  had  passed,  when  it  became 
ammoniacal  and  coloured.  Before  the  ammonia  came  through  a small 
portion  of  the  wet  soil  was  picked  out  from  the  lower  portion  of  the  tube, 
but  it  did  not  smell  at  all  of  ammonia,  which  had,  therefore,  not  yet 
reached  it. 
Another  important  preliminary  inquiry  seemed  to  be,  whether 
a difference  in  the  result  would  be  likely  to  occur  from  a shorter 
or  more  prolonged  exposure  of  the  solution  to  the  action  of  the 
soil.  So  far  as  filtration  was  concerned,  it  seemed  obvious  that  a 
very  rapid  passage  of  the  solution  through  the  soil  being  due  to 
an  imperfect  contact  of  the  two — the  liquid  passing  between  rather 
than  throucjh  the  little  masses — would  fail  to  produce  the  ulti- 
mate effect.  But  it  did  not  appear  whether,  supposing  the  acting 
surface  in  the  most  favourable  state,  the  absorption  would  take 
place  rapidly  and  to  the  full  extent,  or  would  require  a length- 
ened time  for  its  exhibition.  The  following  experiments  were 
made  to  elucidate  this  point : — 
Experiment  8. — A tube  of  24  inches  in  length  was  partial ly  filled  with 
soil,  so  arranged  that  the  liquid  (ammonia  of  -3  per  cent.)  could  be 
forced  through  it  by  means  of  a syringe.  In  this  way  the  solution  was 
made  to  pass  in  something  less  than  two  minutes,  whereas  without  such 
pressure  it  would  have  required  from  one  to  two  hours.  The  first  3 ounces 
that  came  through  were  quite  free  from  ammonia  or  its  salts.  The  ex- 
periment was  not  continued,  as  the  apparatus  would  not  bear  the  pressure. 
Experiments  9 and  10. — Repetitions  of  the  previous  experiment,  with 
precisely  analogous  results. 
Experiment  11.— In  this  experiment  the  soil  was  dried  for  five  or  six 
hours  on  the  sand-bath.  It  was  then  sifted  through  a sieve  of  80  holes  to 
the  inch  to  separate  the  dust,  the  coarser  part  being  employed  in  the  ex- 
periment. The  object  of  drying  and  sifting  in  this  case  was  to  hasten  the 
