an endophytic member of the Erysiphaceae . 19 1 
a rather delicate arachnoid weft. From the sterile hyphae arise the 
conidiophores, which are for the most part simple, though sometimes at 
their base a secondary branch is developed. The conidia, formed at the 
apex of these conidiophores, become abstricted directly they are formed. 
Of these conidia, the terminal one is usually attenuated at the apex and 
truncate at the base, while the others are cylindrical and rounded at both 
ends, and roughly oval. This species is absolutely distinct from the other 
species of the genus [Oidium] in the much greater dimensions of the conidia, 
as well as in the host-plant on which it lives ; it is, however, very like the 
following species [Oidiopsis sicula\ in almost all its characters, although 
the mycelium in the present species is superficial, as in typical species of 
Oidium , while in Oidiopsis siciila it is endophytic.’ 
Through the courtesy of Prof. G. Scalia I have been able to examine 
excellent dried examples of Oidiopsis sicula and Oidium gigasporum , and 
also spirit material of the former. There is no doubt that both belong 
to E. taurica , Lev., and represent the conidial condition of the species 
in the earlier and later stages. In the specimens collected of Oidiopsis sicula 
on Asclepias curassavica which I have seen, no external mycelium is yet 
visible, while in those of Oidium gigasporum on Ballota rupestris mycelial 
hyphae are evident on the surface of the leaf. Representations of the 
fungus on Asclepias are given in Figs. 5-7 and 29. On Ballota rupestris 
the conidiophores are very laxly branched, and very flaccid, and their 
origin from the endophytic mycelium is difficult to trace on account of the 
dense coating of hairs completely covering the under surface of the leaf. 
I have observed clearly, however, that the fungus sends hyphae into the 
stomata (see Fig. 1 8). I have lately received from Prof. F. Bubak exam- 
ples (unnamed) of the same form collected in Montenegro on Ballota acuta 
(— - B. rupestris ), and in these specimens young perithecia of E. taurica are 
visible on the external mycelium among the conidiophores. 
Magnus has recently published (8) a fungus, growing on Asteriscus 
aquaticus (= Odontospermum aquaticum) at Santa Cruz, Tenerifife, as a new 
species under the name of Erysiphe Asterisci l . I have seen the examples 
sent out in Rabenh.-Pazschke, Fungi europ. et extraeurop., ed. nov., ser. 2, 
nr. 4350, and after a careful examination have failed to find any distinguish- 
ing characters from E. taurica. The perithecia, which are semi-immersed 
in the persistent felted mycelium, measure 200-240 p in diameter, and when 
dry become strongly concave or ‘ pezizoid ’ in the manner characteristic of 
E. taurica ; the asci are numerous, and large, and contain two ascospores 
more or less immature. In all respects, in fact, E. Asterisci is typical 
E. taurica as regards the perithecial characters ; and it is somewhat curious 
that Magnus in his diagnosis should have omitted to have compared the 
fungus with this species (especially as Odontospermum had been recorded as 
1 The name (without diagnosis) appears in Engl. Bot. Jahrb. xxxiii, 486 (1903). 
