an endophytic member of the Erysiphaceae. 193 
specimen used by Maire in his examination. I was able to ascertain that 
in these specimens the conidiophores arise from superficial hyphae of the 
mycelium, and that these hyphae send haustoria into the epidermal cells. 
I was not able to observe any endophytic mycelium in the specimens 
I examined. In face of the presence of the normal Oidium- type of 
conidiophore borne by ectoparasitic hyphae, I see no reason to doubt that 
this fungus on Cirsium is E. Cichor ace arum, and I suspect that the 
endophytic hyphae (which Prof. Maire tells me in a letter he observed in 
two or three instances) belonged to some other parasitic fungus. 
I am still of the opinion that E. taurica is distinct generically from the 
other members of the Erysiphaceae in its completely endophytic habit 
during the early part of its conidial stage, and in the production of usually 
branched conidiophores sent up through the stomata. I propose to 
establish for the reception of this genus, which must bear the name 
Oidiopsis , a new Sub-family, Oidiopsideae , with the following characters : — 
Mycelium at first wholly endophytic, producing conidiophores sent up 
through the stomata ; perithecia produced on the hyphae of a superficial 
mycelium originating from the endophytic mycelium. 
The generic characters of Oidiopsis as defined by Scalia will require to 
be considerably extended. Scalia described his genus from the single 
example of the fungus in the conidial condition on Asclepias curassavica, at 
a stage when no superficial mycelium occurred. We have then to add to 
the characters given the presence of the hemi-endophytic mycelium on the 
surface of the leaf (during the later stage of the conidial condition, and 
throughout the perithecial stage), originating from hyphae sent up through 
the stomata, and the generic perithecial characters proper to the fungus in 
its perithecial stage, viz., those of the fungus hitherto known as Erysiphe 
taurica Lev. The genus, then, will stand as Oidiopsis Scalia (emend.), and 
the species as O. taurica (Lev.). 
The Family Erysiphaceae will consist of three Sub-families: 1. 
Erysipheae, in which the mycelium is wholly external to the tissues of the 
host-plant, the hyphae sending haustoria into the epidermal cells alone, or, 
in one species ( Uncinula Salicis (DC.) Wint.), into the sub-epidermal cells 
as well l . Genera : Podosphaera , Sphaerotheca , Uncinula , Microsphaera , 
Erysiphe. 2 . Phyllactinieae, in which the conidiophore and perithecium 
are borne on superficial mycelium, which does not form haustoria in the 
epidermal cells, but sends down through the stomata special branches, of 
limited growth, which send haustoria into cells of the mesophyll-tissue. 
1 genus, Phyllactinia. 3. Oidiopsideae, with the characters given above. 
1 genus, Oidiopsis. 
Thus within the Family Erysiphaceae we pass from ectoparasitism, 
through a hemi-endophytic form, to endoparasitism. 
1 See Grant Smith (13). 
O 
