28  Some  Secojidari/  Actions  of  Manures  upon  the  Soil. 
■ 
have  recorded  that  water  cultures — i.e.,  laboratory  experiments 
in  which  a plant  is  made  to  grow  without  soil  in  water  contain- 
ing small  quantities  of  the  nutrient  salts  it  requires — will 
become  alkaline  in  course  of  time  if  the  solution  is  not  changed. 
To  verify  these  observations  various  water  cultures  were  started 
in  which  vigorous  growth  was  maintained  for  some  months,  at 
the  end  of  which  time  analyses  were  made  both  of  the  plants 
and  of  the  liquid.  The  results  all  confirmed  the  older 
observations  and  the  deductions  that  can  be  made  from 
the  composition  of  the  crops  from  the  field  : the  culture 
solution,  which  represents  the  soil,  became  more  alkaline  as 
growth  proceeds  (or  in  the  majority  of  cases  less  acid, 
because  for  the  success  of  the  water  culture  it  is  desirable  to 
start  with  the  solution  somewhat  acid)  and  the  greater  the 
growth  the  greater  the  amount  of  base  left  in  the  solution. 
In  one  example,  details  of  which  are  quoted  in  the  original 
paper  (loc.  cit.),  wheat  was  grown  in  the  same  jar  of  solution 
from  March  8 until  June  11,  by  which  time  grain  was  fully 
formed  and  the  plants  had  reached  the  weight  of  93-7  grams 
(about  I ounce)  ; by  this  time  the  solution  had  acquired  an 
excess  of  bases  equal  to  about  25  per  cent,  of  the  total  present, 
while  the  plant  contained  a corresponding  excess  of  acids. 
Thus  two  distinct  lines  of  evidence  agreed  in  assigning  the 
production  of  alkali  in  the  soil  to  the  growth  of  the  plant 
itself  ; whenever  a plant  is  fed  with  nitrate  of  soda  some  of 
the  base  will  be  left  behind  in  the  soil  in  the  form  of  carbonate 
of  soda.  Probably  the  nitrate  of  soda  does  not  enter  the  plant’s 
roots  as  a whole,  but  a splitting  iqj  and  selection  takes  place 
at  the  surface  layer  of  the  root-hairs,  where  the  water  and 
dissolved  nutrients  actually  pass  into  the  plant.  At  this  layer 
an  excess  of  the  nitrate  enters,  and  the  soda  that  is  rejected 
thereupon  enters  into  combination  with  the  carbonic  acid 
which  is  always  being  excreted  from  the  same  surface  layer 
of  the  roots. 
Collateral  evidence  is  also  forthcoming  that  nitrate  of  soda 
gives  rise  to  a free  base  in  the  soil — evidence  derived  from  the 
determinations  of  carbonate  of  lime  in  the  Rothamsted  soils, 
to  which  allusion  has  already  been  made.  On  looking  back  to 
Table  I.,  it  will  be  seen  that  the  plots  receiving  nitrate  of  soda 
have  been  losing  their  carbonate  of  lime  less  rapidly  than  the 
unmanured  plots,  one  to  two  hundred  pounds  per  acre  per 
annum  less,  though  too  much  reliance  cannot  be  placed  on  the 
weights  calculated.  This  reduction  in  the  amount  of  carbonate 
of  lime  annually  removed  from  the  soil  is  due  to  the  fact  that 
the  carbonate  of  soda  formed  from  the  nitrate  of  soda  has  done 
part  of  the  work  for  which  carbonate  of  lime  is  usually 
required,  and  has  thus  indirectly  afforded  it  some  protection 
