and  Subsoils  of  some  of  the  Fields  at  Rothamsted.  359 
same  ash-constituents  as  the  four  in  the  o series,  but  with 
200  lbs.  of  ammonium-salts  in  addition,  give  a mean  of  22-7  lbs. 
of  nitrogen  as  nitrates  in  27  inches. 
The  four  plots  in  the  AA  series,  also  receiving  the  same 
ash-constituents  as  above,  but  with  275  lbs.  of  nitrate  of 
sodium,  give  a mean  of  23’9  lbs.  of  nitrogen  as  nitrates  in  27 
inches. 
The  four  plots  in  the  C series,  with  the  same  ash-constituents, 
and  1000  lbs.  of  rape-cake,  give  a mean  of  30T  lbs.  of  nitrogen 
as  nitric  acid  to  the  same  depth. 
There  can  be  no  question  that  the  nitrates  found  in  the  soils  of 
the  O series  have  been  produced  by  the  oxidation  of  the  nitro- 
genous organic  matter  of  the  soil.  In  the  A and  AA  series 
there  can  be  little  doubt  that  a large  portion  of  the  nitrates  has 
a similar  origin  ; Plots  1 and  3 in  these  series  are  the  only  ones 
in  which  it  seems  probable  that  any  residue  of  nitrates  derived 
from  the  manure  can  have  remained  in  the  soil  through  the 
summer.  In  the  plots  of  the  C series,  on  the  other  hand,  a part 
of  the  nitric  acid  is  apparently  due  to  the  rape-cake  applied 
as  manure,  which  only  slowly  decomposes  in  the  soil.  The 
plots  which  have  received  rape-cake  for  30  years  show,  in  fact, 
a higher  average  amount  of  total  nitrogen  in  the  soil  than  any 
other  plots  in  the  field,  with  the  exception  of  those  receiving 
farmyard-manure.  The  considerable  amount  of  nitric  acid  in 
the  plot  manured  with  rape-cake  in  Broad  balk  Field  has  been 
already  noticed. 
There  is  thus,  notwithstanding  the  comparatively  washed- 
out  condition  of  the  plots  when  sampled  after  the  autumn  and 
winter  rains,  still  a sufficiently  marked  difference  between  the 
mean  amount  of  nitrates  found  in  the  soils  of  the  different  series, 
to  be  traceable  to  their  different  conditions  of  manuring,  and 
the  varying  amounts  of  crop  and  crop-residue.  From  the  same 
cause  the  differences  among  the  plots  w ithin  each  series  are  also 
small ; but  there  are  indications  that  these  too  are  dependent  on 
the  varying  conditions  of  manuring  and  growth.  In  series  O, 
A,  and  AA,  the  plots  2 and  4 receiving  superphosphate,  which 
yield  the  largest  crop  and  crop-residue,  are  those  which  generally 
contain  the  largest  quantity  of  nitric  acid.  With  more  growth 
there  would  also  be  more  evaporation ; the  soils  after  harvest 
would  therefore  be  drier,  in  some  cases  considerably  so,  and 
hence  it  would  require  so  much  more  subsequent  rain  to  cause 
the  same  amount  of  passage  of  nitrates  downwards,  and  the 
nitrates  would  accordingly  remain  in  a greater  proportion  within 
the  range  of  the  soil-sampling. 
Among  the  plots  of  the  0 (Rape-cake)  series  a contrary  result 
is  observed.  Plots  1 and  3,  with  somewhat  less  growth,  never- 
