1905] SCH NEIDER—BIOLOGY OF RHIZOBIA 137 
beach vetch the extreme branching form, due to hyper-growth. This 
supposition is strengthened by the study of rhizobia in artificial 
culture media. Grown in the same culture media the rhizobia of 
sweet clover and red clover are morphologically identical. Whether 
they are physiologically identical has not been determined, although 
this is also probable according to the inoculation experiments of 
Nose, HIttNerR, and Horres, of Germany, which are supported 
by the research of Moore of the Department of Agriculture. It 
is quite apparent that these variations in form are due to the food- 
supply or nutritional changes, variations in the supply of oxygen and 
moisture, variations in chemical reaction, temperature, light, and 
other ecological factors. It is possible by means of special culture 
media to augment very decidedly the branching in the rhizobia of 
sweet clover and to induce other morphological variations, as has 
frequently come under my observation. It should be noted that the 
so-called sporoids do not appear in the rhizobia in artificial culture 
media. This is of considerable importance, as it was once believed 
that these bodies were of sufficient significance to be of specific rank. 
It would appear from these observations that they are more likely 
by-products stored within the cell, having perhaps food value, derived 
from the host plant. If this supposition is correct, the theory that 
they are etiologically sporoidal in nature is untenable. 
It seems very probable, and wholly within the range of the possible, 
that the two extreme form types here described are phylogenetically 
derived from an original form type similar to, but not necessarily 
identical with, the form found in Cassia Chamaecrista, Robinia 
pseudacacia, Trigonella joenum graecum, and Amphicarpaea comosa, 
which were formerly described as distinct species. From this it does 
mm follow, of course, that these form types are of necessity varia- 
tions of one and the same species, although the evidence thus far 
deduced points in that direction. 
The presence of the Injectionsfaden was noted (fig. 3), and nothing 
new regarding their significance was discovered, only I wish to state, 
by Way of readjustment of what was stated in previous papers, that 
in all probability these threads are merely a phenomenon dependent 
"pon the infection of the root cells by the rhizobium. The action 
of the motile forms in the apical area of the root nodule (formerly 
