102 
Journal of Agricultural Research 
Vol. XXI, No. » 
The next step was to determine whether this close relation between 
the physical condition of the seed and susceptibility to fungous infection 
holds true of the more natural infections of seed germinating in the soil. 
Spores, were obtained in quantity for soil infection by making cultures 
of Penicillium, isolated from wheat, in 500-cc. flasks containing crushed, 
autoclaved wheat as a medium. Variously injured lots of seed were 
prepared in the same way as those described for blotter germinations and 
were inoculated by soaking about 20 minutes in suspensions of Peni¬ 
cillium spores. The seeds were then sown in pots of sterile, sandy-loam 
soil into which had been mixed quantities of dry spores. After sowing, 
the seeds were watered with the spore suspension. Three sets of four 
pots each, consisting of infected endosperm-injured, embryo-injured, and 
uninjured seed and a control of sterile, uninjured seed were then placed 
in three different temperatures—namely, in the refrigerator, which 
averaged io° C.; in the laboratory, at approximately 20°; and on an outer 
north window ledge having an intermediate temperature. 
Because the growth of wheat is retarded considerably by heavy clay 
soils, a duplicate experiment was conducted in three sets of pots con¬ 
taining Yolo clay loam, in order that every possible chance for infection 
might be given. 
When the se£ds germinated, it was found that only two out of the six 
lots of endosperm-injured seed had escaped infection—those in the heavy 
clay loam in the refrigerator and those on the outer window ledge. Evi¬ 
dently infection was more difficult in this soil, possibly because of in¬ 
sufficient aeration. At the higher temperature of the laboratory, the 
seeds injured over the endosperm, when sown in this soil, produced only 
a low percentage of plants, and these were short and weak. The other 
pots of the clay-loam series, containing the infected, unbroken seeds and 
the radicle-injured seeds, produced plants which were as good as the 
control of uninfected seeds in height, development, and percentage of 
germination. 
The same results were obtained from all three series of pots of the sandy 
loam, only those lots with seed coats broken over the endosperm being 
susceptible to Penicillium infection (PI. 15). All these data are sum¬ 
marized in Table I. 
Table I. —Germination data from a typical experiment to determine the relation of seed - 
coat injury to susceptibility to Penicillium 
Nature of injury. 
Germina¬ 
tion. 
Average 
height of 
seedlings. 
Condition 
Seed coats broken over embryo, 
Per cent. 
96 
Cm. 
15 
Normal. 
infested. 
Seed coats broken over endo¬ 
84 
4 
Weak, with reduced root sys¬ 
tem; seeds covered with 
Penicillium. 
Normal. 
sperm, infested. 
Seed coats uninjured, infested.. 
IOO 
1 5 
Control, uninjured, sterile. 
96 
15 
Do. 
