902 The American Naturalist. [October, 
from near Danzig developed a considerable number of tuber- 
cles in which it was very easy to demonstrate the Y-shaped 
bodies, and from which pure cultures of Nos. 3 and 5 were 
again obtained. Since these two forms behave differently on 
culture media, the author insists that it is no longer a question 
of one tubercle bacillus, but thinks that there are at least two 
and probably more, the form varying with the locality. Water 
cultures were carried on along with the sand cultures, using 
peas and lupines, but with negative results. Some of the roots 
decayed and none developed tubercles. Hellriegel first ad- 
vanced the hypothesis (1886) that the bacteria in these tuber- 
cles are capable of taking nitrogen from the air and turning it 
over to the host plant. This striking hypothesis at once came 
into favor and was accepted as proved by many writers on agri- 
cultural topics. Frank, however, in dry material, found no 
increase whatever of nitrogen when his Rhizobium grew with 
the plants. His many experiments show that the garden bean 
(Phaseolus vulgaris) which always bears tubercles under nat- 
ural conditions never becomes any richer in nitrogen than do 
beans grown in sterile soil and free from tubercles. This cer- 
tainly looks more like parasitism than symbiosis. Other ex- 
periments made by Frank show that lupines and peas can 
assimilate nitrogen when grown in sterile humus, and free 
from tubercles and bacteria. Consequently leguminous plants 
are able to store nitrogen and enrich the soil without the action 
of bacteria, and it is not settled how the nitrogen is taken up 
by the plant. _Gonnermann reasoned that if the bacilli really 
assimilate free nitrogen and turr it over to the host plant, then 
when they are grown in an artificial medium the latter ought 
finally to become somewhat richer in nitrogen. Following 
out this idea, very careful experiments were made with potato 
broth of a known nitrogen content, but although the bacteria 
grew luxuriantly for 14 days there was absolutely no increase 
of nitrogen. The cultures were made in 12 150 ce. flasks and 
every 24 hours the air was changed, being passed through cot- 
ton, strong sulphuric acid, and strong potash liquor to free it 
from dust, microdrganisms, ammonia and carbon dioxide: 
The analyses were made by Dr. Meyer of the Rostock Agricul- 
