151 
which was enriched on an average to 12 milligrams, partly in the 
plants, which had gained 20 to 30 ‘milligrams, although, owing to the 
narrow space in which they were confined, they were not ‘able to 
attain their full development. 
This last proof appears to have finally closed the discussion for- 
merly inaugurated by Boussingault and which had not been com- 
pletely closed by the analytic results explained above. 
Thus a few years have sufficed to definitely decide this theory of a 
direct Foon of nitrogen by plants, first enunciated by Ville. 
What, now, is the mechanism or modus operandi of this assimila- 
tion? We have just seen how Berthelot was led, by certain peculiar- 
ities of his experiments, and, above all, by the complete cessation of 
all fixation of nitrogen in soils that had been subjected to a tempera- 
ture of 100°, to admit that nitrogen is assimilated directly by certain 
inferior organisms which force it into organic combination; but we 
have also seen that the fixation of nitrogen by naked soils is always 
weak and generally insufficient for the necessities of a normal vege- 
tation. 
It is true that when the aid of a leguminous plant is invoked the 
fixation becomes more active and may become powerful enough to 
compensate alone for all the known causes of loss; but how, ‘then, 
are we to account for the difference in this respect found between the 
Leguminose and the Graminee? Shall we be forced to admit that 
the Leguminose are able, by themselves, to assimilate gaseous nitro- 
gen, by a power possessed by them which is wanting in the other 
species 
Berthelot has concluded, from his researches upon this subject, that 
in the development. of leguminous plants there comes into play some 
micro-organism which facilitates the fixation of nitrogen upon the 
root of the plant, or rather upon the mass formed by the root and 
the soil, intimately connected one to the other; but this idea could 
not be definitely adopted unless the existence of such a microbe were 
proved by experiments. This result is fully demonstrated by a 
series of very remarkable experiments made by Hellriegel, Wilfarth, 
Frank, Prazmoftski, and others in Germany, and w hich have been 
most successfully verified by Bréal, Schloesing, jr., and Laurent in 
France, and, finally, by Lawes and Gilbert in England. 
Before proceeding to explain these researches I must call attention 
to-a well-established fact which had been well known for a great 
many years, although no one before Hellriegel and Wilfarth ever 
thought of seeing in it anything more than a phenomenon of nature. 
When we examine the roots of a leguminous plant grown in good 
soil we always see irregularly disposed on them tuberculous enlarge- 
ments, a kind of nodosity [| node, nodule, knot, or knob] formed of a 
special tissue and apparently quite accidental. Examined with a 
microscope the interior of these excrescences appears to be filled w ith 
corpuscles of varying forms, always animated with the “ Brownian ’ 
movement, although they have sometimes a movement of their own. 
These assume various shapes; sometimes they are like simple rods 
similar in form to certain bacteria; sometimes they have the 
appearance of vegetable coralloids and take the branched T or Y 
form more or less ramified. 
Botanists have for a long time discussed the nature of these excres- 
cences, but at present it seems to be generally admitted that, morpho- 
