442 



In the knots and at the top of the short branches 

 (Fig. 7a) round or ovalshaped cells often appear, which 

 by their form give the impression ofconidia; they contain 

 a large quantity of plasmaic matter and are strongly 

 vacuolised. Such cells are also found in the mycélium on 

 the leaves, and when isolated by scratching a withered 

 leaf with a needle, or shredding a pièce of mycélium, 

 they germinate easily. So I suppose that they play an 

 important part in spreading the disease. 



Fig. 8 shows some cases in which new hyphae are 

 originating from such isolated cells. Undoubtedly also the 

 direct contact of diseased and healthy leaves is an impor- 

 tant factor in the process of infecting new branches, for 

 in a moist atmosphère the fungus passes very easily from 

 one leaf to another. I think that leaves may become loose, 

 and be blown by the wind to other trees which they will 

 infect if the conditions are favourable. When dead leaves 

 are hanging from a branch only connected with a mycelium- 

 strand it may often be seen that they touch new healthy 

 leaves and that the hyphae begin to spread from the place 

 where the leaves meet. 



As I hâve already written, I never noticed any fructifi- 

 cation, neither in the natural substratum, or in a damp 

 chamber or any nutritive médium. Of ail the différent 

 média potato-agar gave the best results; besides this I 

 cultivated on sterilised coffee-branches, beanstalks, potato 

 and carrot slices. In coffee-branches the above mentioned 

 knots developed so abundantly, that almost globular bodies 

 with 1 — 2 mM diameter arose; thèse consisted ofarather 

 loose tissue of hyphae. When young they sometimes 

 remind one of the necator stage of djamoer oepas; then 

 their colour is white, but it soon becomes greyish or brown. 

 Thinking that the fungus might belong to the Basidiomy- 

 cetes. I started large cultures on bread with différent 



