Relation to Factor Limiting Bacterial Activity in Soil. 449 



the loss of moisture, and the subsequent rise in the bacterial content may 

 likewise be attributed to the more favourable conditions occasioned by 

 remoistening the soils. 



Considered in relation to their protozoan fauna, these results are very 

 instructive. In the Broadbalk 1856 there are no protozoa. In the Broad- 



20 40 60 68 80 100 102 120 130 140 160 180 198 220 231 



DAYS 



Fig. 5. 



balk 1865 there is a rich protozoan fauna to the extent of about 5000 amoeba? 

 and 5000 flagellates per gramme. In the Geescroft 1865 tliere are about 500 

 amoebae and 500 flagellates per gramme. Thus the curve for the Broadbalk 

 1865 shows that in spite of the presence of this large number of protozoa, the 

 bacteria can maintain a high level in numbers even after 231 days during 

 which the protozoa should have been reducing them. 



It may be suggested that the bacterial counts for Broadbalk 1856 are 

 higher than those for Broadbalk 1865 because in the former there are no 

 protozoa present to check the growth of the bacteria. I would point out, 

 however, that the two Broadbalk curves are practically of the same order as 

 compared with the Geescroft curve. One would have thought that in the 

 presence of such a large number of protozoa as in the Broadbalk 1865, had 

 these been capable of functioning as the limiting factor, they would have 

 checked considerably the multiplication of the bacteria and brought them down 

 to somewhere near the level of the bacterial content of the Geescroft soil. 



My point is to show that the drying to which the Broadbalk soils were 

 submitted has brought about a change in them strictly comparable with the 

 change usually produced by partial sterilisation, and at the same time has 



