16 
The  Production  of  Plant  Food  in  the  Soil. 
partial  sterilisation.  For  a day  or  so  after  the  soil  had  been 
treated  very  little  happened,  then  there  was  a rapid  increase  in 
the  amount  of  ammonia  and  in  the  numbers  of  bacteria  ; the 
nitrates,  however,  did  not  increase  in  the  partially  sterilised  soil, 
and  only  to  a comparatively  small  extent  in  the  untreated 
soil.  The  results  of  one  experiment  are  given  below  : — 
Nitrogen 
Nitrogen  present  ns 
Mlmgen  present  as  ammonia  in 
one  million  parts  of  soil 
nitrate  in 
one  million 
parts  of  soil 
ammonia  and  nitrate 
in  one  million  parts 
of  soil 
At 
After 
Af'er 
After 
After 
At 
After 
At  After  Gain 
begin- 
2 
4 
9 
23 
begin- 
23 
begin-  23  in  23 
ing 
day  b 
days 
days 
days 
ning 
days 
ning  day--  days 
Untreated  soil  
1-8 
20 
22 
25 
17 
12 
16 
138  177  39 
Soil  heated  to  209°  F.  ... 
Soil  treated  with  toluol 
which  afterwards 
65 
" 
97 
281 
438 
13 
12 
]9'5  1 f5  8 363 
evaporated  away 
50 
8’9 
200 
2$'l 
278 
12 
12 
17  0 39  8 22  8 
In  the  twenty-three  days  of  the  experiment  there  had  been 
36-3  parts  of  nitrogenous  plant  food  made  in  the  heated  soil, 
and  22'8  parts  in  the  soil  treated  with  toluol,  but  only  3*9 
parts  in  the  untreated  soil. 
This  is  set  out  as  a curve  in  Fig.  2.  Another  experiment 
may  be  quoted,  showing  the  enormous  increase  in  the  numbers 
of  bacteria  after  the  first  effect  of  the  poison  has  passed  off  : — 
Number  in  millions  of  bacteria  per  gram  of 
dry  soil 
At 
beginning 
After 
10  days 
After 
38  days 
Gain  in 
38  days 
Untreated  soil ... 
4‘2 
10  6 
13-8 
9-6 
Soil  heated  to  209°  F. 
Soil  treated  with  toluol  which 
Very  few 
7-4 
17  6 
17  6 
afterwards  evaporated  away 
1-3 
317 
382 
369 
The  next  step  was  to  discover  the  cause  of  the  increase  in 
amount  of  ammonia  in  the  partially  sterilised  soils.  This 
might  be  due  to  an  increased  production  of  ammonia,  or  to 
the  destruction  of  some  agent  in  the  untreated  soil,  other  than 
nitrifying  organisms,  that  consume  ammonia.  The  second 
supposition  fell  to  the  ground,  because  when  small  quantities 
of  ammonium  salts  were  added  to  untreated  soils  the  whole  of 
the  added  nitrogen  was  recovered  as  ammonia  and  nitrate. 
Hence  we  must  conclude  that  the  treatment  has  induced  an 
