1054 A. F. Vass 
colonies appearing on agar plates must be accounted for in some other way. 
As each colony on a plate may result from a single cell or from a cluster 
of many cells, it would seem that the increase obtained from frozen soils 
might be due to the breaking up of the masses of bacteria, not to an actual 
reproduction. 
Both Conn and Harder considered this breaking-up process to be 
extremely unlikely, for if such was the case the count immediately after the 
thaw would not be so nearly the same as it was before the freeze. This 
objection, however, might not hold in the case of the bacteria, for they 
often grow in a jelly-like mass which might be ruptured by freezing but 
which if allowed to thaw gradually, as it would in the soil under a normal 
condition, might assume its original position. If such was not the case, 
one might expect more bacteria in recently thawed soil than in unfrozen 
soil. If the increase was due to a summer and a winter flora, as sug- 
gested by Conn, there should not be a rapid decrease in number of bacteria 
immediately after the thaw, for the soil would maintain for several days 
a temperature more favorable for the winter than for the summer flora. 
The soils that were allowed to thaw gradually in the field showed approxi- 
mately the same number of bacteria as was shown by the soils before 
they were frozen, whereas the soils that were thawed rapidly in the 
laboratory showed a higher count. This would seem to indicate that the 
rate of thaw might be an important factor. The experiments reported 
herein were conducted with these considerations in mind. 
METHODS OF INVESTIGATION 
ORGANISMS 
In the experiments dealing with the field soils, the bacterial flora of 
the soil was studied as it was observed on examination. A study was 
made of the different groups, but no marked correlation was noted between 
certain groups and the treatment of the soil. Baczllus radicicola from 
field pea was used for all the pure-culture work. This organism was 
selected because of its economical importance and habits of growth. 
It is easy to grow and identify, and is classed as a non-spore-pearing form. 
The latter character is of importance in low-temperature studies. 
MEDIA 
The organisms were grown in soil, sand, and solution cultures. The 
medium used for making the plate counts of the field soils was Conn’s (1914) 
