32 
The Colorado Experiment Station. 
statement, without any experimental evidence, that the cultivation 
of leguminous plants might improve the soil by taking up nutrient 
material from the air and depositing it in the soil by means of the 
roots and stubble. In the light of our present knowledge, this 
was, indeed, a remarkable conjecture. Some years later John 
showed that there was not only an increase in humus following a 
leguminous crop but also a definite increase in the nitrogen con¬ 
tent. This discovery led to numerous attempts to explain the 
source of the added nitrogen. The most common theory advanced 
for this phenomenon was the absorption of nitrogen gas by the 
plants through the leaves, but the classic experiments of Boussin- 
gault, in 1854, together with those of Gilbert, Lawes and Pugh 
showed the fallacy of any such idea. Nothing which approached 
an explanation of the question was reached until 1886 when Hel- 
riegel and Willfarth demonstrated that the source of nitrogen for 
this class of plants was unquestionably the air, and that when 
such plants were grown in nitrogen free soil, the growth always 
took place after the formation of nodules or swellings on the roots, 
and therefore there must be some definite relation between nitrogen 
fixation and root nodules. The results obtained by these two men 
were later substantiated by the investigations of others and a solu¬ 
tion of the long unsolved problem was promised. 
This discovery cleared up quite a big part of the mystery 
connected with the question, and investigators at once turned their 
attention to finding a possible cause for the formation of these galls 
or nodules. Insect bites, worm bites and pathogenic fungi were 
resurrected to explain these hypertrophyed roots. Some consid¬ 
ered them fungus growths similar to sclerotia; others looked upon 
them as modified lenticels, undeveloped buds and spongy roots. 
As early as i860, Woronin described these structures in detail and 
said he believed they were caused by vibrio-like organisms which 
he had discovered within the nodule, but he carried his work no 
farther. Eater observers also found these so-called bacteroids but 
paid little attention to their significance. Accounts of their work 
indicate that they were seeking larger game. 
In 1879, Frank showed that these root tubercles were absent 
on plants grown in sterilized soil, and that they were not mere 
reserve storehouses for protein substances, but stood in direct 
causal relation to the fixation of atmospheric nitrogen. Although 
Woronin almost thirty years before had discovered that these en¬ 
largements were filled with micro-organisms, their true bacterial 
nature was not definitely established until 1888 when Beijerinck 
