210 
seolus vulgaris, P. paucifiorus, Lathyrus odoratus, Pisum sativum 
and Robinia Pseudacacia. According to the statements of Frank 
I expected to find in each species examined the same kind of 
Rhizobium, though perhaps somewhat modified. This, however, 
was not the case. In Melilotus alba, Trifolium repens, T. pra- 
tense and Lathyrus odoratus, 1 found the cells of the infected area 
of tubercles entirely filled with the peculiar Y-shaped organisms, 
(the ‘ Bakteroiden” of Brunchorst, bacteria of Woronin, spores 
or gemmules of Ward, etc.) (fig, 1, plate CXXX.) The same 
tubercles examined in June and July showed the same organ- 
isms, though much changed in appearance. They were larger, 
and the majority took on the Indian-club form. They had be- 
come fattened on the cell protoplasm (fig. 2, plate CXXX). The 
same tubercles examined in September and October showed the 
organisms still more modified. Some of them had become almost 
spherical, others oval-dumb-bell shaped, etc. In some the plant 
had begun to reabsorb the protoplasmic contents, as shown in 4, 
fig. 3, plate CXXX. The same tubercles, examined late in Oc- 
tober and beginning of November, showed that most of them 
had begun to break down, and the contained R/izobia had theit 
entire protoplasmic contents reabsorbed by the plant, leaving A 
nothing but the empty, partially absorbed or decayed walls (fig. : 
3, plate CXXX). In these tubercles were also found a variety of - 
bacteria spores and Rhizobia (?) that were not present while the 
tubercle was intact. Very likely these were foreign to the plant, 
and had been brought to and deposited in the partially destroyed 
tubercle by water currents and capillarity. Probably these for- 
eign spores, bacteria and Rhizobia (?) were the organisms which 7 
Frank took to be the unmodified remnants and spores of the true | 
Rhizobium leguminosarum left behind after the protoplasmic con- - 
tents had been reabsorbed by the plant. Several culture attempts o 
on gelatine in July proved to be failures. This confirms Frank's | 
statements that the modified Rizzobia will not reproduce and . 
develop outside of the host. - Neither by the aid of various stains 
or the use of K H O solution was I enabled to detect the pre 
ence of spores. I am inclined to the belief that they multiply bY = 
direct division only. - 
In the examination of the tubercles of Phaseolus pauciflorus ae 
7 
