EL J BUTTER 5 
occur on grasses, and all of its members are found on these hosts, 
excepting the least studied, 8. Magnusiana Sorokin, which was 
found by Sorokin on Equisetum in Russia. 
Peronospora Maydis Racib. agrees entirely with Sclerospora 
in its asexual stage, differing from all other Peronosporacez in 
the broad, short conidiophores, with short, thick, dichotomously 
formed branches at the top, provided with broadly conical papille 
on which the conidia are formed. The mycelium and haustoria 
also agree, the latter being of the button-shaped type found in the 
stem of Pennisetum. The oogonia differ somewhat, having a not 
very thick, though persistent, membrane. ‘his is not smooth, 
but provided with irregularly arranged, small, conical, warty 
thickenings. In the older descriptions the oogonial wall of Scleros- 
pora was described as fusing with that of the oospore on ripening, 
but Berlese (1902, p. 69) is clearly not of this opinion, and it is easy 
to ascertain from sections or by long maceration that the two coats 
are perfectly distinct. In Raciborski’s species therefore, the se- 
paration of the two walls mentioned by the author does not tell 
against the fungus being a Sclerospora. No doubt the figure 
accompanying the description is about as unlike the oospore of a 
Sclerospora as could be, but it is evidently highly diagramatic, and 
omitting the figure, the description would apply sufficiently to 
Sclerospora. In Berlese’s Monograph it is also suggested that 
the species is a Sclerospora, but a definite opinion is not given there, 
and cannot be given here, in the absence of an examination of 
original specimens. The fungus causes the most intense cereal 
disease produced by the Peronosporacez, epidemic outbreaks in 
Java (where it is known as . ‘‘lijer *’), destroying whole fields of 
maize. It has not so far been reported elsewhere. 
The remaining grass parasites of the genus have been criti- 
cally examined recently by Traverso (1902 (1)). As a result of a 
careful study he gives the following classification, reducing to two 
the three species which had been usually accepted: 
Oospore 28-354. d.; hybernating spore with undulated 
contour, with wall of the oogonium very thick, rubi- 
ginous. Sel. graminicola. 
