244 
Journal of Agricultural Research voi. xxm, No. 4 
on a set of plants extends over two months, then this time will be the 
limit for inducing conidium formation on a similar set of plants which 
after inoculation have been kept indoors except when exposed to dew 
for occasional tests. It should be noted also that the markings of the 
disease on inoculated plants indoors are relatively inconspicuous when 
they appear and gradually become less and less marked until eventually 
they are so greenish in color and so broken up and interrupted in pattern 
that they are only recognized with difficulty. 
In general, conidium production by the downy mildew takes place 
from the surface of the host chiefly on the discolored areas characteristic 
of the disease (PI. 4, B, C, D). It may also occur, although less abun¬ 
dantly, on areas of the host that are apparently normal. Microscopic 
examination shows that the discolored areas harbor the most abundant 
mycelium in their tissues; but even areas that are seemingly healthy, 
and in fact any part of the diseased plant except the roots, may be 
invaded to some extent by the Sclerospora hyphae. As a result, in the 
case of plants which have become infected after they have attained suffi¬ 
cient growth so that they can struggle along to maturity in spite of the 
parasite, abundant conidiophore production may occur not only from 
the leaves but also from the leaf-sheaths (PI. 4, E), from the husks 
PI. 5, B), from the abnormal bract-like structures which are produced 
by diseased tassels and ears (PI. 5, A), from the upper intemodes of the 
stem, from normal parts of the tassel, and from the green, * exposed tips 
of infected ears. 
Conidiophore production may take place on both surfaces of the leaf. 
Whether the conidiophores emerge from the upper or lower surface of a 
leaf depends first, on the presence of stomata through which the conidio¬ 
phores may develop, and, second, on the presence of a sufficient layer of 
moisture to induce this development. In maize the distribution of the 
stomata is rather uniform, the number in the lower surface of the leaf 
exceeding only slightly those in the upper. As a result, if other condi¬ 
tions are equal, the abundance of conidiophore production may be some¬ 
what greater from the lower surface, but as a rule it appears to be about 
the same on both. 
The presence of a layer of moisture on the leaf surfaces is determined 
by climatic conditions. During a gentle, drizzling, nocturnal rain there 
is deposited on the exposed upper surface of the leaf a sufficient layer of 
moisture to induce conidium formation there, while the protected lower 
surface is for the most part dry and bears no conidiophores. Dew, how¬ 
ever, is usually deposited rather uniformly over the infected plant, on 
both the upper and lower surfaces of the leaves, and hence induces 
conidiophore production which either is approximately equal on both 
surfaces or else is somewhat more abundant on the lower, because of the 
more numerous stomata there. 
This relatively even distribution of the conidiophores induced by dew 
is the type most frequently encountered on maize in the Philippines, 
and the reports of Butler (5), Rutgers (. 20 ), and Palm (16) that conidio¬ 
phores occur on both surfaces but are slightly more abundant on the 
lower would indicate a similar condition in the case of the other oriental 
maize mildews. On sugar cane, however, at least on the only 
conidiophore-producing plant which the writer found in the Philippines, 
the conidiophores are almost wholly confined to the under surface of 
the leaves, and this is true also of the wild grass Saccharum spontaneum. 
