1894. ] Mutualistic Symbiosis. 29 
Culture experiments were made with special precautions to 
prevent the introduction of bacteria, etc., everywhere present 
on the surface of both tubercles and roots. As a suitable 
medium a slightly acid agar-vegetable root extract was pre- 
pared. Tubercles ofnormal appearance were secured and care- 
fully and thoroughly washed with plenty of hydrant water 
then quickly dried by means of blotting paper which had been 
passed through the flame of a Bunsen burner. A tangential 
section, including somewhat more than the dermal layer, was 
made down one side of the tubercle, then the tubercle was 
passed through the flame of the Bunsen burner so as to singe 
it thoroughly on all sides, and then broken (not cut) across. 
Inoculations were made from the broken surface farthest from 
the cut side. The inoculated tubes were placed in a dark 
chamber at the ordinary summer temperature (Ills.). In about 
six or seven days a small whitish growth was noticed in 
most tubes. Cultures made from the dermal and hypoder- 
mal parenchyma generally developed organisms resembling 
Rhizobium Frankiit (plate Iv, fig. 6), and I shall provisionally 
place them with that genus. I lati de from the palisade 
layer sometimes developed a coccus, more often there was no 
growth at all. As arule there are no bacteria or cocci to be 
found with the infecting nostoc. Cultures made from the 
vascular system, especially near the apical area, generally 
developed the above mentioned Rhizobium, but more often a 
peculiar Indian club shaped organism (plate Iv, fig. 8). I was 
unable to obtain absolutely pure cultures, but the two forms 
of bacteria (or rhizobia) described seemed to predominate. Cul- 
tures in which the rhizobium-like organism predominated 
finally took on a yellowish color. In this respect it resem- 
bled very much cultures of the rhizobia from the ‘‘Infektions- 
fiden” of Melilotus alba or T; rifolium pratense. This organ- 
ism resembles in appearance Rhizobium Frankii of Phaseo- 
lus vulgaris but differs in that it has cilia and is motile dur- 
ing its earlier life history. Rhizobia and bacteria are not 
Present in large numbers in any part of the tubercle. The 
question whether their presence is purely accidental or 
whether they live in active mutualistic symbiosis with Cycas 
could not be determined in the short time at my disposal. It 
's however quite certain that tubercles contain more bacteria 
TE ereetirreinsereesin gg 
eae eens 
a * * S . . 
. “Purther experiments in regard to the determination of the rhizobia are now 
10 progress. 
