124 CO. CLIFFORD DOBELL. 
organism—for instance, in the case of the spirilla and 
spirochets. In speaking of Bacillus bitschl1i, Schaudinn 
says, “ Biegung des Stibchens findet man ausserst selten ;” 
so that in this respect it must differ markedly from my 
Bacillus. The only other organisms which appear to be in 
any way comparable, are the remarkable forms included in 
the Myxobacteriaceee of Thaxter. But here, although the 
rods are flexible, they are not motile in the same manner, for 
they progress by creeping movements on a resistant surface. 
(See Baur’s description.) 
Division is effected in B. flexilis by a simple constriction 
into two, as in B. sporonema, Schaudinn. In the method of 
division, therefore, another point of difference between my 
organism and B. biitschlii is seen: for in the latter, division 
takes place by septation. I have watched division in the 
living organism, and have stained permanent preparations at 
various stages, but I think it is not necessary to enlarge upon 
a process which has been already described in so many bac- 
teria (see Plate 6, fig. 10). 
The most interesting phenomena are those connected with 
spore formation. I will describe these first, and comment on 
them afterwards. Unfortunately, I have not been able to 
follow the whole process from beginning to end in one and 
the same living organism: for, at the end of a period varying 
from three to five hours, my protists always died in hanging 
drop or coverslip preparations. But I have been able to 
watch spore-formation in different organisms in consecutive 
stages, and I believe, the following is a substantially correct 
account of the normal course of events. The observations 
have all been confirmed on stained material. 
The organisms which are about to form spores are of con- 
siderable length—usually about 30 to 35 u—and are filled 
with distinct granules, which stain a purple-red with Giemsa. 
After swimming about in this condition for some time, the 
rod begins to show a slight constriction at its middle, and 
appears to be about to divide (fig. 10). Division is not com- 
pleted, however, and after the lapse of some time, it can be 
