S SOME DISEASES OF CEREALS. 
result of insect or fungus attack. Many instances of the latter are 
mentioned by Tubeuf and Smith (1897, p. 26) and Rostrup (1886). 
Thus Peronospora violacea causes the stamens of Knautia arven- 
sis to become petaloid. Ustilago violacea causes the rudimentary 
stamens of the pistillate flowers of Lychnis dioica to develop fully. 
The reverse occurs under the influence of several Ustilaginee, 
ovaries developing in staminate flowers, as in Carex preecox infect- 
ed with Ustilago Caricis. Physoderma deformans Rostr. causes 
virescence and malformation of the flowers of Anemone nemorosa. 
The extraordinary hypertrophies and distortions of the flowers of 
Crucifere attacked by Cystopus candidus are well known. Mol- 
liard (1901) has even shown that parasites acting at a distance can 
produce these modifications ; as, for instance, in Primula officinalis, 
where double flowers result from the attacks of a Dematium on the 
roots ; and petalody of the stamens of Scabiosa Columbaria was 
experimentally induced by him by inoculation of the roots with 
Heterodera radicicola. 
In grasses, several instances of floral modifications of this 
nature are known, the most important of which are caused by 
fungus parasites of the genus Sclerospora, one of the Peronos- 
poracee. It is to one of these that the disease just described 
is due. 
The mycelium of the parasite is found in all parts of the 
attacked plants which show the alterations mentioned above. 
Thus it has been found in the stem, leaves, rachilla, bristles of the 
involucre, glumes and prolificated shoots. 
The hyphe are large, unseptate, very variable in size, up to 
nearly 104 in diameter, and run ‘chiefly intercellularly (plate IV, 
fig. 4). They are found in the ground tissues of the stem, and the 
mesophyll of the leaves chiefly, but branches penetrate the bundle 
sheaths in the leaves, where they lie between the sheath cells, and 
others collect under the stomata and in young leaves send out clus- 
ters of condiophores through these into the air. Only in rare cases, 
chiefly in late stages, have hyphe been observed in the inner walls 
of the epidermal cells sending haustoria into them. The 
xylem and phloem elements proper never contain the fungus, 
