350  Nitrojen  a?  Nitric  Acid,  in  the  Soils 
is  recorded,  the  pipes,  in  fact,  did  not  run.  In  all  these  cases 
loss  of  nitrates  by  drainage  and  diffusion  doubtless  took  place, 
though  we  have  no  means  of  estimating  the  quantity  actually 
passing  below  the  first  27  inches  of  soil.  Evidence  that  the 
amount  passing  into  the  subsoil  by  drainage  and  diffusion  may 
in  some  cases  considerably  exceed  that  estimated  from  the  com- 
position of  the  pipe-drainage,  is  afforded  by  results  obtained 
relating  to  the  amount  of  the  chlorides  present  in  the  soil. 
As  a very  considerable  amount  of  drainage  took  place  in 
Broadbalk  Field  after  the  samples  of  soil  were  taken,  it  is 
obviously  of  interest  to  compare  the  results  of  the  analyses  of 
the  soils  with  those  of  the  subsequent  drainage-waters.  If  the 
drainage-waters  fairly  represent  the  condition  of  the  first 
27  inches  of  soil  on  the  different  plots,  the  quantities  of  nitrate 
found  in  the  waters  should  show  a similar  relation  between 
themselves,  as  exists  between  the  nitrates  found  in  the  soils  of 
the  corresponding  plots.  This  comparison  of  the  results  of  the 
soil-analysis  with  the  composition  of  the  subsequent  drainage 
from  the  same  plot,  indeed  brings  to  view  some  facts  which 
would  otherwise  escape  attention. 
Turning  now  to  the  details  of  the  soil-analyses,  we  observe, 
first,  that  the  nitrates  are  most  abundant  in  the  first  9 inches  of 
depth  ; the  mean  proportion  at  the  three  depths  (excluding 
Plot  9b)  is,  in  fact,  as  100,  59,  and  31.  The  nitrates  have, 
in  fact,  mainly  been  produced  near  the  surface,  in  most  cases 
quite  recently,  and  are  now  gradually  passing  downwards.  The 
case  is  quite  different,  however,  on  Plot  9b,  where  more  nitrate 
of  sodium  has  been  applied  than  the  crop  could  make  use  of; 
here  the  nitrates  are  present  in  largest  quantity  in  the  second 
and  third  depths,  and  a part  has  probably  passed  to  a still 
lower  level.  A large  portion  of  this  nitrate  has  existed  in  the 
soil  ever  since  the  application  of  the  manure  in  March,  and 
has  since  then,  and  especially  since  the  removal  of  the  crop, 
been  moving  slowly  downwards  by  diffusion  and  drainage. 
From  a similar  cause,  the  quantity  of  nitrate  in  the  lowest  depth 
of  Plots  8a,  9a,  and  10a  is  also  above  the  average. 
Passing  next  to  the  quantity  of  nitrate  found  in  the  various 
plots,  we  observe  that  the  unmanured  soils  of  Plots  3,  4,  and  16a, 
yield  the  lowest  amounts,  namely  17*8  lbs.,  15  0 lbs.,  and  17*9 
ibs.  of  nitrogen  as  nitrates  per  acre.  Plot  5a,  continuously 
manured  with  ash-constituents,  yields  24-3  lbs.  ; this  amount, 
however,  bears  a higher  relation  to  that  contained  in  the 
subsequent  drainage  than  is  shown  by  the  adjoining  Plots  3,  4, 
6a,  and  7a.  Plot  17a,  also  receiving  ash-constituents  alone 
during  the  year  in  question,  yields  21*2  lbs.  per  acre.  Of  the 
effect  of  the  various  ash-constituents  in  promoting  nitrification 
