THE VINE-MILDEW. 269 



boloid only occurred, though Erysiphe lamprocarpa (on Salvia 

 glutinosa, Sonchus oleraceus, Plantago major), Er. communis (on 

 Trifoliura pratense, Ranunculus acris) developed both forms close 

 to each other. Under these circumstances it is explicable that 

 I found only the Erysiphoid fruit in Podosphaeria Kunzei, 

 Phyllactinia guttata, Uncinula Wallrothii, U. Bivonae, Erysiphe 

 Martii, E. horridula, while on the contrary in a series of forms on 

 Brassica oleracea, Heracleum Sphondylium, Cucurbita Pepo, 

 Pyrus Malus, Viola tricolor, there was a rich growth of Cicino- 

 boloid fruit, but not a single Erysiphoid, on which account I could 

 not determine to what particular genera or species they belonged. 



In order to determine the relations of the two forms of fruit to 

 each other, it will be sufficient to examine the fructification of 

 Sphgerotheca Castagnei. The upright threads of the fungus 

 which eventually produce conidia capable of germination are 

 generally divided by dissepiments into a greater or less number 

 of cells (8 — 11) of which the lower are cylindrical, the upper oval. 



When one of these threads is changed into a Cicinoboloid 

 fruit (or as it is termed by Tulasne, pycnidium) the change from 

 cylindrical to oval does not take place, but on the contrary one 

 (or sometimes two) of the lower cells swells out and is metamor- 

 phosed, in consequence of the formation of a layer of small yellow 

 cells on the inner side of the wall of the cell, into a cellular sac 

 whose cavity is stuffed with minute ovate spores (stylospores of 

 Tulasne), and at its apex the unchanged portion of the thread 

 remains as a long articulated white appendage. The form of this 

 kind of fruit is not determinate, and when the cells enter into its 

 composition the relative size is subject to considerable variation. 

 When mature it bursts irregularly at the apex and protrudes a 

 mass of spores in the form of a tendril mixed with a gummy 

 substance which remains unchanged in alcohol, but dissolves in 

 water and breaks up into the individual spores. 



I was not able to detect accurately the formation of the 

 Erysiphoid fruit from a single cell. I feel quite sure about the 

 matter after much investigation, that the notion of Ehrenberg 

 (Act. He. Nat. Cur. x. 205) that it springs from the union of 

 many threads of the mycelium is incorrect, but inasmuch as it 

 adheres closely to the horizontal threads from which it springs, or 

 only on a very short ramification, the observation of its first 

 development is surrounded with great difficulties ; I believe, 

 however, that I have not deceived myself in thinking that I have 

 seen it proceeding from a single thread, which at an early period 



