536 
Journal of Agricultural Research 
Vol. XXXI, No. 6 
at the hilum and had formed the usual dark-brown intercellular 
aggregates of mycelium in that part of the parenchyma of the middle 
layer of the seed coat which remains uncollapsed in the hilum region 
of the seed (pi. 5, D, G). These mycelial aggregates were conspicuous 
immediately around the fibrovascular bundle which extends from the 
funiculus into this tissue. Sclerotial bodies were also formed in this 
tissue (pi. 5, E, F, G). In certain instances observed the mycelium 
had invaded the parenchyma of the middle layer as far as the 
boundary of the endosperm (pi. 5, G), and in some cases had even 
grown down a short distance at the sides of the endosperm, forming 
dense aggregates and occasionally sclerotial bodies in this region. 
With the collapse of the middle layer, these bodies are crowded 
between the endosperm and the outer seed coat and cause the latter 
to bulge outward at that point (pi. 5, E, F). No mycelium was 
detected within the endosperm, however, probably because of the 
apparent absence of intercellular spaces. 
In addition to the internal infection of the seeds, there was also 
a more or less extensive mycelial infestation of the palisade layer 
of long cellulose rods or hairs which cover the outer integument of 
the seed. The mycelium in some cases had accumulated in these 
nairs near the hilum and formed well developed sclerotial or peri- 
theciumlike bodies entangled among the hairs and thus firmly 
attached to the seed coat (pi. 5, H). Thus it seems that the seed 
may be infested by the fungus externally as well as internally. 
The identity of the fungus within and on the seeds was proved 
by incubating the latter in a moist chamber. Eleven seeds which 
showed infected hila and which had been dried 10 days were placed 
on moist filter paper in a Petri dish. Abundant sporulation of 
Cladosporium occurred on the blackened tissues of 6 of these seeds 
within 8 to 13 days (pi. 5, B). Long spore chains were produced, 
some containing as many as five spores. The spores were typical of 
C. fulvum and prompt germination occurred in water, in fact, 
many of the spores had germinated in the damp air of the Petri dish. 
Agar plates poured from a suspension of these spores yielded typical 
O. fulvum colonies. 
Leaf infection of young tomato plants was obtained by spraying 
with a suspension of the spores from one of the cultures isolated 
from a seed. Sporulating lesions were found 23 days afterwards on 
all of the inoculated plants, while the check plants remained healthy. 
One plant was successfully inoculated by spraying the lower sides of 
the leaves with water and touching some of the drops with a seed 
coat bearing conidia about the hilum. 
In one of the cases of successful sepal inoculation previously de¬ 
scribed, the fruit was examined 14 weeks after the inoculation and it 
was found that abundant seed infection had occurred. 
SEED TRANSMISSION 
That the mycelium of Cladosporium fulvum may invade the seed 
and establish itself on and within the seed coat has been shown. 
In no case does it appear that the endosperm or embryo is actually 
invaded by the fungus. The parasite may also gain access to the 
seed in another way. It was found that in extracting the seed from 
diseased fruits fragments of the blackened, infected fruit tissue 
very frequently adhere to the surface of the seed, and that, after 
