209 
cocci. That all RAzzodia may have a similar function, that of assi- 
milating free nitrogen, does not indicate that there must be but 
one species. Even since Frank maintains that the Bakteroiden 
are true, though modified, R/zzobia he seems to believe that there 
is but one species. In 1890 he declared that the micrococci-like 
Rhizobia were contained within the Bakteroiden which, according 
to his recent conclusions in regard to Bakteroiden, would be ab- 
surd or at least improbable. 
Frank also seems to take it for granted that the plant or rather 
root has some attractive influence on the R/zzobia. He gives no 
cause for this assumption, nor do experiments bear him out. 
Capillarity, wind and water are the principal agents for the dis- 
tribution of micro-organisms in soil. Air currents carry them 
from place to place on the surface of the soil. Rain, for instance, 
carries them deeper into the soil and brings them in contact with 
roots and rootlets; these with their root hairs and probably a 
sticky secretion, arrest them and infection takes place at suitable 
points. Capillarity will either carry them down or up, that 
depending upon the condition of the soil as to moisture. Hell- 
riegel and Beyerinck have shown that when a leguminous plant 
was planted in sterilized soil and one side sprinkled with normal 
culture soil solution and the other with sterilized culture solution, 
the former developed tubercles while the latter did not. Now if 
the roots had any attractive influence they would certainly have 
drawn some of the Rhizobia from the infected to the non-infected 
Side, 
The purpose of my observations was primarily to settle the 
following questions : 
(1). Do all the tubercles of the Leguminose contain one and 
the same micro-organism ? 
(2). Do all the micro-organisms undergo a mo 
form in all infected Leguminose ? 
(3). Is the micro-organism a Rhizobium or a Ehylornye ? 
In May, 1891, I examined the tubercles of 7: rifolium doesent 
and 7, repens and noted the appearance of the contained Rhizobia. 
They were as shown in fig. 1, plate CKXX. Later, in June and 
July I examined tubercles of Amphicarpaea comosa, —— 
Chamecrista, Trifolium repens, T. pratense, Melilotus alba, Pha- 
dification of 
