102 BACTERIA IN RELATION TO PLANT DISEASES. 
The effect of the soil-inoculation on legumes differed from the effect of nitrate of soda in that, 
in the former case after the period of germination, a peculiar and very characteristic hunger-stage 
supervened which was followed by very energetic and rapid development of the plants. 
. In two experiments under sterile conditions, the peas grew well in the nitrogen-free sand until 
the food stored in the seed was exhausted and then dwindled, dying after about 6 leaves had been 
formed. On these plants not a trace of root-nodules could be found. The same negative result was 
obtained when the soil-extract was boiled or heated to 70° C. before adding it to the pots. 
‘They concluded, therefore, that the nitrogen assimilation of legumes was in some way connected 
with root-nodules and the bacteria present therein. 
In other words, as expressed in their final report: 
There are, therefore, for the Leguminosae two sources of nitrogen, viz., the combined nitrogen 
of the soil and the elementary nitrogen of the air, the latter being made available to them through 
the agency of micro-organisms which, to be effective, must enter into a symbiosis with the plant. 
Numerous experiments with lupins failed: No successful second growth could be obtained with 
pea soil and the conclusion was reached that the nodule organism of lupins must be different from 
that of peas. Only when inoculations were made with soil from a field where lupins grew well 
did the experimental plants overcome their nitrogen-hunger and do well. On this experiment two 
check rows of pots were held, one untreated and one inoculated with extract from pea soil. In all 
three the plants germinated and grew well at first. Then followed a period of starvation, each of the 
three rows showing equal nitrogen-hunger at the end of a month. Then the first row became green 
and grew well, while the other two rows continued feeble and red-brown in color. The roots of the 
first row (inoculated with lupin soil) bore numerous large nodules. The roots of the second row bore 
none whatever. The roots of the third row (inoculated with pea soil) bore none whatever, except 
one plant on which asingle small nodule wasfound. Serradella behaved like the lupin. Peas, vetches, 
and beans grew best in the third row. 
In Hellriegel’s own words: 
“‘Leguminosenknéllchen und Wachsthum der Papillionaceen in stickstofffreiem Boden lassen 
sich willkiirlich hervorrufen durch Zusatz von geringen Mengen Kulturboden und Verhindern durch 
Ausschluss von Mikroorganismen. Bei verschiedenen Papillionaceenarten wirkt nur der Zusatz von 
gewissen Bodenarten Knéllchen bildend und Wachsthum férdernd.”’ 
Lawes and Gilbert sum up these experiments very well in the following paragraph: 
“The negative result with the Gramineae, the negative result with the peas when everything 
was sterilized, or when the sand was not seeded by the soil-extract, the positive result with the peas 
when the sand was seeded by the humus soil extract, the negative result with the lupins when their 
soils were not seeded, or when they were seeded with the same extract as the peas, and the positive 
result when seeded with the extract from the sandy soil where lupins were growing, seem to exclude 
any otherconclusion than that the micro-organisms supplied by the soil-extracts were essential agents 
in the process of fixation. Further, the development of nodules on the roots was, to say the least, 
a coincident of the fixation.’’ 
The following year (1887) Dr. Wilfarth stated at the Naturforscher Versammlung in Wiesbaden 
that they had repeated and extended their experiments with wholly confirmatory results (plate 12). 
From this time on the scientific world generally accepted their views as may be seen from the fol- 
lowing comments of Lawes and Gilbert: 
‘Thus it may be considered established that the Papillionaceae can take the whole of their 
nitrogen from the air. * * * 
“Tt will be seen that the results are not only confirmatory of those given by Hellriegel the year 
before, but that they are even much more definite and striking. Thus, taking no account of the 
fraction of a milligram of combined nitrogen supplied in the soil-extract, the amount of dry matter 
produced is nearly 50 times, and the amount of nitrogen assimilated is nearly 100 times, as much 
with, as without, the soil-extract.’’ 
The full account of their experiments was first published in 1888. No figures since published 
are any more striking or convincing than the six plates which appeared in this epoch-making pub- 
lication, two of which are here reproduced (plate 12 and fig. 32). 
In their experiments with serradella, Series C, 1897 (plate 12), each jar contained 4,000 grams 
of sterilized nitrogen-free quartz sand to which was added the necessary nitrogen-free nutrient 
salts (monopotassium phosphate, potassium chloride, calcium chloride, and magnesium sulphate). 
Eight seeds were germinated in each jar, the number of plants being reduced soon after to four. 
The plants were watered with distilled nitrogen-free water. 
To some jars additional fertilizers were added as follows: Nos. 264 and 265 each a small amount 
of calcium nitrate (41 mgs.); the two end pots of each row (right side) received each 40 grams of 
calcium carbonate which was sterilized by heat and mixed with the sand previous to planting; No. 
250 received some potassium carbonate. 
