MICHIGAN ACADEMY OF SCIENCE. 
49 
faster and more complete decomposition in soil cannot be explained 
by drainage water, removing the harmful products and thus giving bac- 
teria a. new chance for activity. Drainage water contains very little 
organic matter which is of a nature that it could not possibly cause any 
retardation of microbial action. 
This leads to another problem which is perhaps the most prominent 
one of today, namely the difference of microbial action in soil and in 
solution. It seems strange that until a few years ago soil bacteriologists 
paid no attention whatever to the physical structure of soil and car- 
ried on all experiments in solution. The dairy bacteriologist grows 
his organisms on milk or milk products, the veterinary bacteriologist 
grows bacteria for experimental purposes in the animal body, while the 
soil bacteriologist does not nse soil as a basis for his soil experiments, 
but meat media and occasionally soil extract. It is very true that there 
is not much difference chemically between the so-called soil solution 
which is the natural habitat of soil bacteria and between a soil extract. 
Not the chemical qualities, nor the amount of food make the great dif- 
ference between soil and soil solution, but the physical nature of the 
two. Bacteria live only in liquids. In soils, the liquid is spread in a 
thin ‘film over all the finest soil particles, and thus the surface is many 
hundred times enlarged. Doubtless, other soil qualities, like absorp- 
tion, influence microbial development, but no other factor has such an 
enormous influence as the exposure of a large surface of liquid to the 
air in a well-aerated soil. The abundant supply of oxygen is a great 
stimulus for aerobic bacteria, while anaerobic organisms are decidely 
suppressed. This can be illustrated by a few experiments in which 
quartz sand has been used instead of soil in order to eliminate the 
possible influences from any organic or inorganic soil compounds. A 
peptone solution was sterilized in a flask and inoculated with a bacter- 
ium that formed ammonia from peptone. The same amount of the same 
solution was mixed with some pure sand, sterilized and inoculated with 
the same pure culture. After 7 days, ammonia was determined in both 
cultures, and the sand culture had formed three times more ammonia 
than the ordinary liquid culture from the same amount of peptone. 
With two other bacteria, five times and eight times more ammonia was 
formed in the sand cultures. Evidently these bacteria required plenty 
of oxygen, since the large supply of air caused an increase of ammonia 
production of 200 to 700%. A similar experiment was carried on with 
the nitrogen-fixing Azotobaeter, which gave in sugar solution 4.2 mgs., 
in sugar solution plus sand 35.0 mgs. of nitrogen, the increase caused 
by sand being 720%. While aerobic bacteria are thus stimulated by the 
change from liquids to soil, anaerobic organisms are retarded in their 
development. Bacterium lactis aeidi was found to make only about half 
as much acid in milk, if quartz sand was added to the milk. Similar is 
the experience with denitrifying bacteria which were feared so much 
because they destroy nitrates. They are feared no more, for the ex- 
periments leading to their discovery were not properly conducted. 
They are found in most soils and they will destroy nitrate if the 
soil is submerged in a solution, but they will not attack nitrates at all 
in a well-aerated soil. Koch and Pettit showed recently that in soil 
with a fair moisture content, nitrates are not destroyed because the 
