4 BULLETIN 1029, U. S. DEPARTMENT OF AGRICULTURE. 
Histological studies to trace the course of the mycelium within 
and beneath the seed coat have not been made. It became evident, 
however, as a result of cultural studies, that the effect on diseased 
seeds may vary from little or no change in normal color and little 
shriveling to very marked shrinkage and entire loss of vitality, de- 
pendent upon the extent of invasion by the fungus. That the infec- 
tion may be confined to the seed coat is shown by one instance where 
a seed which had been treated with 1 : 1,000 mercuric chlorid to kill 
organisms on the surface was placed on a potato-agar plate. Normal 
germination occurred and the hypocotyl as it developed became 
attached to the agar about half an inch away from the seed coat. 
As growth proceeded the young seedling was drawn entirely away 
from contact with the testa and was later removed to a tube of agar, 
where it remained sterile. Subsequently, however, a pure culture of 
Phoma lingam developed from the seed coat, proving that the fungus 
was present in this portion of the seed without having invaded the 
cotyledons. 
In the study of the effect of fungicides upon germination and 
subsequent development of seedlings from normal cabbage seed, most 
of the work was done with two lots (Nos. 2-18 and 3-18) of the 
Wisconsin Hollander variety grown at Racine, Wis., in 1918. In 
general, one lot (No. 2-18) showed more injury from disinfectants 
than the other, in spite of the fact that a high percentage of the 
untreated seeds germinated and produced vigorous seedlings. This 
may have been due to improper curing, since the seed plants after 
being harvested were piled indoors before the seed was thrashed and 
the circulation of air provided was insufficient for rapid and thor- 
ough curing. A portion of lot No. 2-18 was stored in a cloth sack 
in the laboratory, and certain of the treatments were repeated in the 
second year (winter of 1919-20). After the general limits of injury 
with these two lots of seed had been determined, the effect of stronger 
treatments was usually extended to several other lots of various ages 
and from different sources. 
In a few of the early experiments germination tests were made in 
Petri dishes lined with moistened filter paper. It was found that 
numerous treated seeds often gave evidence of germination by throw- 
ing off the testa and starting growth of the hypocotyl, but develop- 
ment ceased at this point. When parallel tests were made in soil in 
the greenhouse, most of such seedlings never appeared above ground. 
Consequently, in order to secure a more definite index of the capacity 
of various lots of seed for producing vigorous seedlings after treat- 
ment, the Petri dish method was discarded. All results reported on 
the effect of fungicides upon commercial seeds are from tests made 
in silt-loam soil in the greenhouse. The final estimate of the amount 
of injury was supplemented in the case of the stronger treatments 
