458 FETCH : 



Diplodias. their separation is doubtful. Diplodiopsis P. Henn. , 

 if it has no more distinctive characters than " superficial, 

 granulato-rugulose," cannot be se]pskva,tedivoii\Botryodiplodia ; 

 both characters may be found in the specimens of Botryo- 

 diplodia theobromce winch develop from the points of origin 

 of the lateral roots of young Hevea. 



Diplodia Wurthii Koorders appears to furnish a further 



example in support of the above conclusions. Tliis species 



was discovered on leaves and stems of Ficus elastica in Java. 



Its pycnidia are scattered or gregarious,. sometimes confluent, 



150-250 [J. diameter ; the spores measure 22-30 X 12-18 jx, 



and it has paraphyses 50-60 \i long. Koorders states (11) 



that the pycnidia which develop on nutrient media exhibit a 



development of hairs at the apex in some cases. He relates 



how, when a dead leaf of Ficus elastica wliich bore several 



pycnidia was left wrapped up in a damp cloth for several 



days, numerous pycnidia clothed with a tliick growth of hairs 



developed on the under surface. These hairs developed from 



the apex and the upper part of the wall of the pycnidia, as well 



as from a black stroma, and formed a web of hyphse ; they 



originated beneath the epidermis, but subsequently broke 



through and became cliiefly superficial. Koorders notes that 



if the non-pilose form were not known, these specimens might 



have been referred to Lasiodiplodia or Chcetodiplodia. But 



since the black web of mycehum and the remarkable stroma- 



like structure of the pycnidia are only the result of the 



changed external conditions, he considers that the fungus 



should be referred to Diplodia. The behaviour of tliis species 



is identical with that of Botryodiplodia theobromce. It must, 



however, be pointed out that in aU probabiHty this example 



does not reaUy afford additional evidence in favom' of the 



general conclusion that Lasiodiplodia and Chcetodiplodia are 



founded on inconstant characters, for it is in the liighest degree 



probable that D. Wurthii is identical with Botryodiplodia 



theobromce. The latter grows abmidantly on dead Ficu^ 



elastica in Ceylon. 



As we have already seen, the presence of paraphyses in the 

 hymenium has never been considered a character of sufficient 

 importance to warrant the separation in distinct subgenera 



