OKDER ERYSIPHALES (PERISPORIALES OF MOST AUTHORS) 311 



Erysiphe graminis DC. may consist of 20 or even more before the older 

 ones break off. Berlese (1898) reports for this species that after a cell is 

 cut off at the top of the conidiophore it divides into two conidia, the next 

 cell cut off from the conidiophore dividing similarly into two conidia. Thus 

 the terminal two conidia are of equal age and are the oldest pair, followed 

 by younger and younger pairs toward the base of the conidial chain. The 

 conidia are distributed by the wind and germinate on the epidermis of 

 the hosts, producing a short hypha which sends a haustorium into an 

 epidermal cell. Conidia are not described for the genus Astomella (Thiru- 

 malachar, 1947). (Figs. 97, 98.) 



Brodie (1945) summarizes the experiments of himself and others upon 

 the germination of the conidia of Erysiphaceae at different degrees of 

 relative humidity. The conidia of Sphaerotheca pannosa (Wallr.) Lev. var. 

 rosae Wor. will not germinate at a relative humidity below 95 per cent. On 

 the contrary a considerable percentage of the conidia of Erysiphe graminis 

 DC, E. polygoni DC, and Microsphaera alni (DC.) Wint. germinate in 

 air entirely devoid of moisture, i.e., zero per cent relative humidity. This 

 accounts, at least in part, for the frequently observed phenomenon that 

 some powdery mildews appear abundantly in very dry weather. Possibly 

 also the dryness of the atmosphere keeps down the infection by Ciccin- 

 ohohts cesatii de By. and other fungi that are parasitic upon these mildews. 

 As long as the chains of conidia remain attached to their conidiophores 

 none of the conidia germinate but when detached the terminal conidia at 

 each end of the chain will germinate, but not those between. 



After infection has taken place the mycelium grows rapidly, branching 

 in all directions, but gradually spreading radially. Conidial production 

 begins soon and may continue for some time but eventually gives way 

 to the production of perithecia. In some species of powdery mildews 

 conidia only are produced on some hosts, conidia and, later, perithecia 

 developing on other hosts. When perithecial development occurs it is 

 often on only one side of the leaf although conidia may be produced on 

 both sides. This is not universally true. Perithecial production is rare in 

 the tropics. 



We owe our first clear knowledge of the sexual processes in this family 

 to Harper (1895 and later) although de Bary and others 20 to 30 years 

 earlier had described the external features. From neighboring hyphae in 

 contact there arise an antheridial and an oogonial branch, each at first 

 uninucleate. The latter is somewhat rounder and the former a little more 

 slender. They become appressed side by side or may even coil about each 

 other a little. The antheridial branch divides into a uninucleate stipe and 

 an apical uninucleate antherid which presses closely against the upper 

 portion of the oogone. The oogonial branch remains unicellular or may 

 divide into a smaller basal and a larger terminal cell (the oogone proper) 



