Jan. 27 , 1923 
Production of Conidia in Philippine Sclerosporas 257 
Moreover, such early-infected plants are usually much more liable to 
the destructive attacks of such secondary parasites as Pythium, Fusarium, 
Helminthosporium, and bacteria. 
Occasionally the period of conidium production is prolonged by the 
abnormal growth of the diseased plant. In response to some stimulus 
exerted by the parasite, various parts of the host, especially the shank, 
husks, ear, and tassel, may develop abnormally, the activity of the 
mycelium keeping pace with this unduly extended growth of the plant 
body. In one striking case of this sort, a plant of Guam maize (PI. 5, B) 
which showed symptoms of infection after vigorous growth had well 
begun, supported nightly conidium production by the downy mildew 
from the time it was 17 days old during its subsequent development. 
After about 5 weeks, when a normal tassel had been produced, the very 
young ear began to develop; and, abnormally stimulated by the downy 
mildew mycelium which had invaded the bud as it formed, the shank 
grew excessively in length, forming a branch which was clothed with 
extensive leafy husks, almost equaled the main axis in length, and bore 
at its tip a small sterile ear. The remarkable growth of this abnormal 
ear branch occupied about 30 days. The branch was still green and 
vigorous long after the healthy plants nearby were dry, and about 14 
days after the upper part of the main axis had begun to wither. During 
all this time nightly conidium formation by the parasite was taking 
place from all parts of the excessively developed leafy husks of the 
abnormal branch, thus prolonging the period of conidium production 
by the infected plant over a total of more than nine weeks. Finally, 
about two weeks after conidium formation was observed for the last 
time, the vitality of the branch declined and it gradually withered. 
The behavior of the downy mildew on its other hosts presents an 
interesting comparison to that which is characteristic for maize. In the 
few sorghum plants that became infected, the period of conidium pro¬ 
duction was necessarily brief, as death occurred a few weeks after infec¬ 
tion. This was true also in the only cases of conidially infected sugar 
cane which were encountered (25). Moreover, on both these hosts the 
nocturnal conidium production during this brief period was far less 
extensive and abundant than in maize. In teosinte, on the contrary, 
the duration of conidiophore production is prolonged beyond that of 
maize, because of the constant renewal of young tissue suitable for spore 
production. As new shoots are sent out from the base of the infected 
plant (PI. 6, A, C), or as new branches arise, the fungus mycelium of 
the main axis grows out into the freshly formed tissue, keeping pace in 
its development with the renewed growth of the plant, on which it 
produces additional conidiophores. In this way conidium production, 
although not so abundant on any one night as in maize of equal age, is 
prolonged and continues constantly and persistently for several months. 
This was also true in teosinte-maize hybrids, which showed extensive 
growth of numerous suckers and branches. In Miscanthus japonicus the 
period of conidium production was similarly prolonged. It is in Sac - 
charum spontaneum , however, that this prolongation is greatest. Infected 
plants kept under observation were still producing conidia on many newly 
formed shoots even after an uninterrupted period of over eight months 
(25). Moreover, conidium production was renewed on new shoots 
arising from the well-established rootstocks of infected plants even 
