Feb. io, 1923 
Life History of Azotobacter 
405 
Inspection of figure 1 of the preliminary paper (28) will show that the 
increase in the number of growth types from four to seven is due to the 
fact that types I and G (with K7) could also be stablized, and that the 
three types E, Fa, and F0 could be grown separately. 
In accordance with the arrangement made in Part I of these Studies 
{23) the different cell types, the various modes of reproduction, the sym- 
plastic stage, and the conjunction of vegetative and reproductive cells will 
be discussed consecutively. The cultural characteristics of the different 
developmental stages, however, will be given separately in a special 
chapter. The morphological features of vegetative and reproductive cells 
were found to be fundamentally the same with all our Azotobacter strains, 
while the cultural peculiarities allow a clear separation of the four species 
(or varieties) Azotobacter chroococcum , A . Beijerinckii , A . agile, and A. 
vitreum. Accordingly, the morphological facts observed with the different 
strains will be considered jointly, while the cultural characteristics of the 
four groups will be discussed separately. 
I.—DIFFERENT CELL FORMS OF AZOTOBACTER 
The wide pleomorphism of Azotobacter was discussed to some extent 
in earlier publications (j, 9, 26 , 30 , and 40). But how really bewildering 
the multitude of cell types is which this organism is able to produce in the 
course of its life cycle may be gathered from the photographs reproduced 
on the plates attached to this paper. Even they, of course, do not ex¬ 
hibit all forms which were observed in the course of our studies. But 
combined with the data presented in our preliminary report (28) and in 
Part I (2 5) they will suffice to indicate that an approximately complete 
picture of the form cycle of a bacterium is very different from the cus¬ 
tomary meager description of a so-called bacterial species. 
The relatively large globules, ovals, or rodlike cells, measuring about 
2 to 4 by 3 to 5 to 7ju, motile either (ovals) by polar or (rods) by peri- 
trichous flagella, mostly Gram-negative, are in most cases very unstable. 
Figures 1, 2, 4, and 6 to 8 of Plate 1 demonstrate how even in young 
cultures (2 to 7 days old) the typical large cells are inclined to change to 
small globular forms (coccoids) or to more slender pointed rods, character¬ 
istic of the sporulating type of growth. In older cultures large and small 
rods, irregular fungoid, as well as very small dwarfed cells are always 
present, as is shown in figures 3, 5, 9, 11, and 12 from growths 3 weeks 
to 11 months old. All these forms are viable, and therefore not to be 
classed as involution forms. Figures 1 to 5, 9, and 11 make it fairly 
clear that the decrease in cell size is largely due to a loss of voluminous 
slimy cell elements and to the liberation and multiplication of nuclear 
material, which according to the quality of accessory substances presents 
itself as either well or weakly stainable, or as entirely unstainable when 
treated with aqueous dyes. Potato agar favored especially the ap¬ 
pearance of globular cells; slightly acid mannite-nitrate agar (P H 6.0) 
strengthened the tendency to produce small rods and coccoids. Only 
two of our cultures (No. 13 and 18) grew permanently on all substrates 
in the large typical form, but also in these cases the microscope always 
revealed the presence of the other cell types in small number. 
The varying appearance and behavior of the coccoid cells of medium 
and of small size is illustrated on Plate 2. Figures 13 to 15 demonstrate 
their budding out of the larger cells; figure 16 shows multiplication by 
budding and by fission; figures 17, 18, and 21 represent the fairly uniform 
