Mar. 2o, 1916 Relation of Green Manures to Failure of Seedlings 
1171 
unsterilized; B, 1.5 per cent of green manure, sterilized; C, 1.5 per cent of 
green manure, unsterilized; D , 3 per cent of green manure, sterilized; 
E } 3 per cent of green manure, unsterilized. The soil shown in the pots 
in Plate LXXXIII, figure 5, was treated with green oats, in Plate 
TXXXIV, figure 6, with green clover. Since the corn and wheat did 
not show any injury, these illustrations were not reproduced. The data 
from cotton, clover, and flax are presented in Plate LXXXIV, figures 1, 
2, 3, 4, and 5. A glance at the seedlings in the sterilized soil shows con¬ 
clusively that heat removes or renders inactive the harmful factor. The 
percentage germination of all crops in the sterilized green-manure soil 
was equal to that of the untreated controls. Apparently, sterilization 
has in some way prevented any injury from green-manuring. This is 
true with 1.5 or 3 per cent of green manure. When repeated, the same 
results were obtained. These data are given in Table XIII. All of the 
results point to an injurious agent of biological nature. 
Table; XIII .—Effect of heat on the germination of cottonseed 
Letter. 
Treatment. 
Germination. 
Relative. 
1 week. 
2 weeks. 
3 weeks. 
Per cent . 
Per cent. 
Per cent. 
Per cent . 
A 
None. 
95 
IOO 
IOO 
IOO 
B 
Sterilized.: 
8 $ 
•85 
8S 
8S 
c 
1 per cent of clover. 
IO 
IO 
IO 
D 
1 per cent of clover sterilized. 
80 
80 
80 
80 
E 
i per cent of oats. 
35 
35 
35 
35 
F 
1 per cent of oats sterilized... 
85 
90 
90 
90 
• 
SOURCE OF INJURIOUS AGENT 
When portions of diseased seedlings are used to inoculate sterilized 
green-manured soil, the germination of oil seeds is greatly reduced. 
Numerous tests show that the harmful agent is readily transferred. 
From the data it must be concluded that the injury to seed germination 
is biological, probably due to bacteria or fungi. To study the nature 
of the agent, a series of tests was made with different micro-organisms. 
EFFECT OF BACTERIA 
In this series of tests bacteria from seed, from green manure, and from 
soil were studied. From the nature of the seed coat of cotton it is no 
doubt very rich in a number of bacteria. According to plate counts, 
the number of micro-organisms on cottonseed is over 122,000 per gram, 
or an average of nearly 11,000 organisms to one seed. A comparison 
of the germination of cottonseed free of bacteria and with bacteria, in 
unsterilized green-manured soil, did not disclose any difference in germina¬ 
tion. The bacteria were removed (2) by exposing the seed to the action 
of hot mercuric chlorid (HgCl 2 ) or concentrated sulphuric acid (H 2 S 0 4 ). 
The use of sulphuric acid offers an easy and satisfactory method of 
