Feb. io, 1913 
Life History of Azotobacter 
423 
and it is also our experience that this feature is rather unstable even in 
B. asterosporus , as is proved by the permanent absence of ridged star 
spores in the original culture, obtained from the New York Museum of 
Natural History. 
When the strains show a somewhat more vigorous growth and their 
cells increase in size, the morphological and cultural characters become 
identical to those of Bacillus fusiformis A. M. et Gottheil, according to 
the descriptions published by Gottheil ( 12) and by Lawrence and Ford 
(jo), with the exception that not only globular spores were produced but 
also many of oval shape. Comparative tests made with a culture from 
the New York Museum gave identical results. Only on potato agar 
nearly all spores were globular, and these showed a pronounced inclination 
to become normal regenerative bodies. Apparently B. centrosporus Ford 
(17) should also be classed as a variety of B. fusiformis. The cultural 
marks are very similar, and the particular form of the rods is equally 
noticeable with otherwise typical B. fusiformis growth (fig. 63 on pi. 6). 
The next step in the development of the small sporulating rods of 
Azotobacter chroococcum and A. Beijerinckii proved to be a change in 
cell morphology from the slender, frequently pointed forms of Bacillus 
fusiformis to a more compact appearance with rounded or square ends, 
centrally located spores, and a change in the cultural character to that of 
Bacillus pumilus A. M. et Gottheil. Ford was unable to obtain a typical 
culture of this strain, and the one we received from Krai's Museum also did 
not fully agree with Gottheil’s description; it displayed in several respects, 
especially on potato, unmistakably the characters of Bacillus fusiformis , 
that is, while our cultures gaining in size and vigor of growth changed 
from the type of Bacillus fusiformis to that of Bacillus pumilus , the 
weakened stock culture of the original isolation had followed the opposite 
course. Bacillus pumilus in its typical form makes a very characteristic 
grayish pinkish brownish smooth growth on potato, well known to the 
senior author from numerous isolations made under his direction by 
Bierema (6 ) in 1908. Two of these cultures were still at our disposal, as 
was also one culture of Bacillus Freudenreichii (Miquel) Mig., described 
by the senior author in 1905 (23, p. 719) and later recognized as closely 
related to Bacillus pumilus (6). The cultural features were still fairly 
similar to those of the original isolations, but the spore-formation had 
ceased entirely, and the cells exhibited clearly a mixture of fungoid and 
small rod-like growth. As was to be expected from these findings further 
testing confirmed that this fungoid growth of Bacillus pumilus is identical 
with the white-pink Mycobacterium developed from A. chroococcum and 
A. Beijerinckii as well as from Bacterium lactis viscosum. It was not 
difficult to change these spore-free strains of Bacillus pumilus and 
Bacillus Freudenreichii by continued cultivation in milk into typical 
strains of Bacterium lactis viscosum. On the other hand we have not yet 
succeeded in reestablishing the endosporulation, which once was very 
conspicuous with these strains, whose purity and authenticity are be¬ 
yond doubt. Successive transfers on potato and heating in beef broth 
up to 85° C. clearly favored the return to a more typical rod form (fig. 
58, 54, and 46 on Plates 5 and 4) and also to the production of fairly 
resistant zygospores and arthrospores. But a complete success has not 
been reached thus far, and this experience makes it easy to understand 
why analogous tests with Azotobacter have been and are so often failures. 
Still somewhat larger than Bacillus pumilus is a sporulating rod 
which was developed directly from the dwarfed growth (No. 2 grown 
