The  Production  of  Plant  Food  in  the  Soil.  1 1 
forms — (1)  in  the  humus,  (2)  in  the  non-humic  part,  (3)  as 
ammonia,  (4)  as  gaseous  nitrogen.  The  relative  amounts  of 
ammonia  and  of  gaseous  nitrogen  depend  on  the  conditions. 
If  the  soil  is  very  rich  in  organic  matter  a gi  eat  deal  of  free 
nitrogen  is  given  off  ; thus  on  the  Rothamsted  plots  receiving 
14  tons  of  dung  annually  no  less  than  half  of  the  added 
nitrogen  is  lost  as  gas  ; while  rich  prairie  soils,  when  first 
broken  up,  suffer  heavy  losses  of  this  element.  From  the 
cultivator’s  point  of  view  this  part  of  the  process  is  sheer 
waste  ; it  sets  a limit  to  the  richness  of  arable  soils  and  to  the 
usefulness  of  heavy  dressings  of  dung.  On  the  other  hand,  if 
the  soil  is  poorer  in  organic  matter,  containing  only  those 
quantities  present  in  ordinary  arable  land,  a smaller  proportion 
of  the  nitrogen  is  lost  and  more  of  it  appears  as  ammonia.  On 
the  Rothamsted  plots  receiving  dung  once  in  four  years  nearly 
70  per  cent,  of  the  nitrogen  passes  through  the  ammonia  stage. 
Ammonia,  however,  does  not  remain  as  such  in  the  soil. 
The  instant  it  is  formed  it  is  attacked  by  certain  bacteria  and 
changed  into  nitrites,  which  in  turn  are  attacked  by  other 
bacteria  and  changed  to  nitrates.  This  last  is  the  most  rapid 
in  the  whole  chain  of  processes ; the  formation  of  nitrite  is  also 
quicker  than  that  of  ammonia  : hence  we  never  find  nitrites 
or  anything  more  than  a trace  of  ammonia  in  ordinary  soils. 
One  of  the  steps  previous  to  the  ammonia  stage  takes  place 
more  slowly,  and  so  sets  a limit  to  the  speed  of  the  whole 
series  of  changes. 
There  is  another  important  process  quite  distinct  from  the 
breaking  down  of  organic  matter,  although  closely  associated 
with  it.  Certain  bacteria  have  the  power  of  absorbing  gaseous 
nitrogen  and  using  it  as  food.  This  is  the  most  mysterious 
process  of  all,  and  cannot  by  any  artificial  means  be  reproduced 
under  soil  conditions.  Some  of  the  organisms  are  associated 
with  leguminous  plants  and  cause  part  of  the  enrichment  of 
the  soil  produced  by  the  growth  of  clover,  tares,  &c.  Others  can 
work  on  their  own  account  in  the  soil  provided  there  is  an 
ample  supply  of  organic  matter  of  the  right  kind,  and  these 
probably  contribute  to  the  gain  in  nitrogen  in  arable  land 
laid  down  to  grass. 
The  importance  of  these  changes  in  promoting  soil  fertility 
is  obvious.  The  ley,  the  stubble,  and  a large  part  of  the  dung 
ploughed  in  are  themselves  of  no  use  as  plant  food,  nor  do 
they  help  to  make  a tilth.  They  open  up  the  soil,  which  may 
be  an  advantage  in  some  cases,  but  by  no  means  always. 
But  the  humus  to  which  they  give  rise  is  always  valuable  ; it 
binds  a loose  soil,  lightens  a heavy  one,  and  increases  the 
power  of  holding  water.  The  ammonia  and  nitrates  produced 
by  the  micro-organisms  furnish  the  nitrogenous  food  for 
