114 Proceedings of Indiana Academy of Science. 
naturally arose as to whether the counts obtained in this and the pre- 
vious work were not due to increased bacterial activities as the ground 
thawed. 
To give more definite information, special experiments have been 
carried out. A special bacteriologist’s soil sampler*® reinforced with steel 
was secured and driven down into solidly frozen soil. The sampler con- 
taining the frozen soil was brought into the warm laboratory and in a 
half hour it was possible to push the core of frozen soil out of the 
sampler. This core was placed on a laboratory table. A wire was 
pushed into the core from time to time and it was found that thawing 
took place very slowly. It was forty-six hours from the time that the 
sample was laid out on the table before it had thawed enough for the 
wire to be thrust through it. 
To see if the bacterial numbers in soil were not increased on the 
thawing out of the soil due to different layers of the soil being brought 
successively under more favorable conditions for bacterial development, 
the following test was made: 
A sample of frozen loam soil was obtained, brought to the laboratory, 
pushed out of the sampler, then taken to a room having a temperature 
below 0° C., where it was halved lengthwise by chopping with an axe. 
One-half was chopped and mixed and fifty grams weighed out and 
analyzed immediately for its bacterial content. The other half was 
brought to the laboratory and allowed to stand twenty-four hours. It 
thawed out in this time. The sample was mixed and its bacterial con- 
tent determined. The results of this test were that the sample allowed 
to thaw out before it was analyzed gave over three times the bacterial 
count that the one analyzed immediately did. 
The following experiment is the latest one we have conducted on 
this subject, and it is left to the reader to judge from this in connection 
with the other work reported as to whether bacteria multiply in frozen 
soil. About twenty kilos of soil (silt loam) were procured by taking 
soil from between the depths of four and seven inches of a plot where 
millet had been plowed under each of the two preceding springs. This 
soil was mixed and sieved through a screen having eight meshes to the 
inch. The portion passing the screen was mixed thorovghly and then 
quartered. One quarter (about five kilos) was brought to the labora- 
tory. Sterile 12-ounce bottles plugged with cotton had been previously 
prepared and 150 grams of the mixed and prepared sample were weighed 
out into each of twenty-six bottles. The soil in the bottles was then 
compacted by dropping them on the bench thirty times. The bottles 
were then divided into three grcups and these groups were incubated in 
the following places: 
