NOTES ON THE ERYSIPHACEAE 19 
tung zu sein. Perithecienbildung setzt einen aus Alteren (meist 
ausgewachsenen) Pflanzentheilen bestehenden und durch Conidien- 
fructification noch nicht erschopften Nahrboden, sowie ein mehr 
oder weniger reich entwickeltes Luftmycel voraus.”’ 
I cannot agree with Neger that he has here given the causes 
which account for the formation or non-formation of perithecia in 
the Erysiphaceae. We find in nature so very frequently perithecia 
produced on the same leaf by the mycelium which had previously 
produced conidia that I am inclined to look upon this as the 
normal course of development, and to consider the non-pro- 
duction of perithecia on conidia-bearing mycelium as exceptional, 
instead of vice versa, as we should be obliged to do if we accepted 
Neger’s view. In many cases, the non-production of perithecia 
seems undoubtedly connected with the influence of certain host- 
plants on the fungus ; ¢. g., itis only in rare cases (see monograph, 
p. 202) that Erysiphe cichoracearum produces perithecia on culti- 
vated plants of Cucumis and Cucurbita. Further, it is to be noted 
that when perithecia do occur, they are found (in the case of the 
fungus on Cucurbita Pepo (vegetable marrow), not on distinct parts 
of the host-plant, but in the midst of the parts covered with the 
Oidium-bearing mycelium. Exactly the same occurs in the ex- 
amples of £. cichoracearum on the tobacco plant. On several 
species of Myosotis an Oidium occurs which has been commonly 
referred to £. cichoracearum, but neither on living plants, nor on 
the numerous specimens in exsiccati and herbaria, have I ever 
been able to find any perithecia. In conformity with the rule 
which obtains for many parasitic fungi, the normal course of de- 
velopment for the Erysiphaceae consists of two phases—a conidial 
stage in which successive crops of conidia are produced during the 
summer months when the host-plant is growing vigorously, fol- 
lowed (omitting exceptional cases) by a formation of perithecia at 
the time when the life of the host-plant begins to fail, so that rest- 
ing-spores (here represented by the ascospores are formed in order 
to continue the life of the fungus in a dormant condition through 
the winter when no host-plants are available. This relation to 
food-supply may be compared with the well-known fact of the 
production in the Zygomycetes of zygospores, which can be arti- 
ficially induced at any time by diminishing the food supply of the 
fungus. 
