36 
MILDEWS, RUSTS AND SMUTS 
plants. The entire group is often spoken of as powdery 
mildews, on account of the dense masses of conidia that are 
produced, and rest on the white patches of mycelium, 
giving them the appearance of having been s])rinkled with 
flour. These conidia are readily dispersed by wind, insects, 
etc., and infect adjoining plants. The conidial condition 
is well developed and constant in all species, but as a rule 
the conidia are so very similar in the different species, that 
they are of little or no value for the purpose of determining 
species, and it is only by infection experiments that their 
true nature can be ascertained, as they will only infect 
their own proper host-plants. The minute perithecia, 
at first yellow, then blackish, are readily seen with a pocket- 
lens, on the surface of the leaf resting on the superficial my¬ 
celium. Sections of the leaf, at those points where the my¬ 
celium is located, will be necessary to show the structure 
of the haustoria in the epidermal cells. The perithecia 
should be picked off the leaf, and placed in a drop of water 
on a slide, and covered with a cover-glass, when the 
character of the appendages can be studied, and if the 
perithecia are gently crushed, the asci are pressed out, 
when their shape, size, and number of spores contained, can 
be studied. 
At one time the connexion between the white, conidial 
condition of the fungus, and the ascigerous stage 
was not suspected, and the white conidial stage was con¬ 
sidered as a genus by itself, and was known as Oidium. 
Perhaps no other family of fungi, until the appearance 
of Salmon’s exhaustive monograph, was in such a state of 
hopeless confusion, as proved by the synonyms, which are 
almost legion. The determination of species, or the crea¬ 
tion of new ones, being apparently determined more by the 
host-plant, than by morphological characters presented by 
the fungus itself. Nevertheless, we find that if a given 
species of fungus is not confined to a single host-plant, it 
is at least generally confined to allied host-plants, in which 
case the host-plant at least gives a clue to the species of 
fungus present, this clue, however, must not be accepted 
as final, and the fungus itself must be examined before its 
name can be correctly determined. 
ERYSIPHACEAE 
Parasitic on living plants ; mycelium slender, colourless ; 
conidial form of reproduction, along with the mycelium 
produced on the surface of the host-plant; ascigerous form 
