98 STORRS AGRICULTURAL EXPERIMENT STATION. 
In determining these groups the primary question that arises, 
of course, is what characters shall be used to separate the 
groups. While all characteristics should, of course, be taken 
into account, some are clearly of more importance than others. 
There is among bacteriologists as yet no consensus of opinion 
as to what characters are of most importance for this purpose. 
It is, however, quite generally thought that morphological 
data are primary and that these should be the first points of 
distinction between these organisms. We are inclined to be- 
lieve that this is true, but as intimated above, we have been 
forced to think that it is necessary to make certain modifica- 
tions of this statement. The cocci and bacteria certainly run 
into each other in such a way that it is sometimes simply im- 
possible to determine whether a culture should be called an 
extremely short bacterium or a coccus type. As already stated, 
the same culture will be described. by one person as one and by 
another as the other of these two forms. Moreover, the classi- 
fication of the spherical forms as usually accepted to-day into 
the streptococcus and micrococcus groups appears to be imprac- 
ticable in many cases. We have found many forms of bacteria 
in which it is practically impossible to determine whether we 
are dealing with a coccus that divides in one plane only or in 
two planes. Moreover, in one or two cases we have clearly 
found a spherical form that multiplies for a long time by divid- 
ing in one plane, producing a long chain, which would be 
then naturally called a streptococcus, and then the whole series 
of spheres divided in the other plane, at once giving chains of 
pairs. This observation has been made once or twice in our 
own laboratory and has been confirmed by Winslow. Whether 
to call such a type a streptococcus or a micrococcus is evidently 
a problem. We have furthermore found in our study that 
many forms which are described as streptococci are identical in 
every character with others described as micrococci except this 
one point of the method of division. These facts have con- 
vinced us that this distinction of the micrococci and strepto- 
cocci is an uncertain one, and we have not found it possible to 
follow it out with accuracy. We have therefore used this 
character as a secondary rather than a primary one in distin- 
guishing types. The question of flagella present upon micro- 
organisms has appeared to us to be of more importance than 
