Jan. 27,1923 
Production of Conidia in Philippine Sclerosporas 265 
tions showed that the spread of the downy mildew in the corn plantings 
closely followed this air flow in direction and extent. Plots containing 
recently germinated and extremely susceptible maize seedlings developed 
few if any cases of the downy mildew when located toward the mountain 
to windward of actively infective plants. Similar plots, however, situ¬ 
ated away from the mountain to the leeward of the same actively in¬ 
fective plants and in the direct path of the stream of conidia borne on 
the nightly air currents from them developed a very high percentage 
of downy mildew cases. 
In larger fields it was possible also to demonstrate that the spread of 
the downy mildew closely followed the direction and extent of the nightly 
air currents. When, after carefully removing any actively infective 
plants so situated as to be dangerous to such a field, a single conidium- 
producing plant was transplanted to the center of it, that portion of the 
field located to windward of the diseased plant remained free from disease 
except in the very immediate vicinity of the plant itself. In the leeward 
part of the field, however, as the young com developed, there subsequently 
appeared numerous cases of downy mildew. These were especially 
frequent along the fan-shaped path of the conidia which were carried 
by the nightly air currents from the centrally placed diseased plant 
and were increasingly numerous the nearer one approached the source of 
infection. Although, as has been mentioned, the general trend of these 
nightly air currents was down from the mountain, it was found by experi¬ 
menting that their extent and direction were greatly influenced by local 
conditions. Such topographical features as wooded ridges, hills, deep 
gullies or brook beds, and small bodies of water had a marked effect; 
while buildings, thick woods, and dense growths of tall cogonal grasses 
often served also to obstruct or deflect the air currents. Even trees, 
bamboo or banana clumps, and small plots of sugar cane, sorghum, or 
cassava were found to disturb the nightly air currents by causing little 
eddies and local circulations of air. 
Petri plates containing a thin film of clear 1 per cent agar were employed 
to gain a more exact idea of the direction and extent of conidium dis¬ 
persal by these nightly air currents. On an apparently still night such 
plates were exposed during the period of maximum conidium formation 
at various distances around a single diseased plant which was producing 
abundant conidia. All other actively infective plants in the vicinity 
having been removed, careful record was made both of the position of the 
plates and of the direction of the prevailing air currents in relation to the 
diseased plant. At dawn the plates were carefully examined under the 
microscope and the presence and number of conidia noted. Very few 
spores were found in plates situated to windward of the infective plant, 
and only in those which had been very near it (within 3 or 4 feet). To 
the leeward of the infective plant, however, and directly in the path of 
spore distribution especially, large numbers of conidia were caught. 
These were most numerous in plates near the plant and proportionately 
less abundant in those at greater distances. Plates within 8 or 10 feet 
were liberally showered; those 20 and 30 feet distant showed many 
conidia; while those situated 50, 60, and in one case, even 80 feet away 
still caught a few. As the Petri dishes were of small size (10 cm. diameter) 
it was to be expected that at greater distances the conidia would be so 
scattered as to reduce to the minimum the chances of their being cap¬ 
tured. It should be noted that the conidia whether caught at 8 or 
80 feet germinated vigorously on the moist agar surface. 
