272 
Journal of Agricultural Research voi. xxm, n 0 . 4 
Mycelium-infested seed, in contrast to cuttings, apparently plays no 
part in the distribution of Philippine maize mildews. Although in maize 
the writer (24) has found Sclerospora mycelium in the seed coats and even 
invading the endosperm, a series of experimental plantings of such seed 
in all stages of development, and even with the diseased cob still 
attached, gave rise to no mildewed plants. Similar plantings of such 
seed from infected plants of teosinte (Euchlaena luxurious Schrad.), 
teosinte-maize hybrids, and Saccharum spontaneum gave like negative 
results. In the Javan maize Sclerospora, Palm (16 ), after careful 
histologic study and extensive plantings, found the same to be true. 
These facts are of interest, as it has long been assumed that seed- 
borne mycelium is a means by which Sclerospora is distributed. This 
assumption has been based on investigations carried out on Sclerospora 
macrospora Sacc. in which the oogonial phase alone is known, and on 
S. graminicola , in which the conidial stage is rare and the oogonial 
preponderates. It should be noted that in the cases recorded infection 
might well have arisen, not from seed-borne mycelium but from oospores 
which in both species are produced in vast numbers and were undoubtedly 
adhering to or present in the coats of the seed used. This explanation 
would apply to Peglion’s observation (17) that grain from wheat heads 
infected with 5 . macrospora and showing mycelium in the outer coats 
produced, when germinated, abnormal seedlings with mycelium in their 
tissues. It would also apply to the case described by Massee ( 14) of an 
outbreak of Sclerospora in England on sugar-cane grown from seed from 
France, a case that is particularly interesting not only because it 
presents evidence for seed transmission, but also because the resulting 
downy mildew was S. graminicola , a species as yet unrecorded on cane. 
DISSEMINATION BY MEANS OF OOSPORES 
When the distribution of the Philippine Sclerosporas by oospores is 
considered it is impossible to state how much or how little is accom¬ 
plished by these bodies, since unfortunately the exact role which they 
play is not at present understood. That they must be functional to 
some extent in enabling the disease to withstand unfavorable conditions 
or to spread over distances so great that conidia would be killed in 
transit seems logical and necessary. Yet, in so far as the writer was 
able to determine, oospores are never produced on maize in the Philip¬ 
pines. The conidial stage alone occurs on this host, persisting, although 
somewhat diminished, through unfavorable seasons, spreading disas¬ 
trously in favorable periods, and causing widespread destruction through¬ 
out the islands, all apparently quite independent of any oogonial phase. 
So far as their existence on maize is concerned one need consider the 
Philippine Sclerosporas only in the conidial stage to understand their 
persistence, their spread, and their destructiveness. But, on the other 
hand, on other gramineous hosts in the Philippines it is the oogonial 
stages that are predominately widespread, persistent, and destructive. 
On Saccharum spontaneum , for example, Sclerospora oogonia are ex¬ 
tremely abundant, not only in the lowlands where the conidial phase, 
although rare on this host, is common on maize, but also in the moun¬ 
tainous interior of Luzon where conidia on either host do not occur or 
are so rare as to escape observation. Similarly, in this same moun¬ 
tainous region the wild grass Miscanthus japonicus and the cultivated 
sugar-cane show abundant infection with oogonia exclusively. Moreover, 
