18 
Further (p. 305) “the ripe fruit of the sweet-chestnut. 
(Castanea vesca, Gaertn.) is sometimes filled with a dense 
blackish-olive felt belonging to the fungus named above. Noth- 
ing as to the life history of the parasite is known.” 
On comparing these descriptions one is at once struck 
by the fact that Berkeley makes no mention of this characteristic 
blackish-olive colour in his careful description of the fungus. 
This fact puzzled me for some time, for the fungus, as I found 
it and cultivated it, was obviously enough Acrospetra mirabilis, 
yet it showed no signs of this colouring. Further acquaintance 
with the fungi found in chestnuts has led me to conclude that 
Massee’s description refers in part to one of the Dematiee 
(which at present I cannot get to fruit), frequently occurring in 
them and traversing the whole nut with its olive-black hyphe.* 
Further, his figure in the British Fungus Flora, Vol. III, 
p. 358, fig.. 13, does not agree with Berkeley's orginal descrip- 
tion or figures, or yet with a specimen from Broome in the 
Cambridge University Herbarium. It is suggestive of an 
imperfect realization of the true meaning of Berkeley's account. 
The chestnuts from which I obtained my supply of Acrospeira 
were bought in Cambridge. The only information I was able 
to obtain with them was that they were grown in all probability 
in the South of Spain, and that the disease in some seasons 
caused a considerable loss. This was clearly the case, for 
thirty-five out of fifty nuts first examined were found to contain 
the chocolate-coloured powder formed by the loose mass of 
spores. As the nuts were too dry for the fungus to have spread 
much during transit, I assume they were infected before 
importation. The spores often filled the space between the 
cotyledons, and more rarely were found in abundance 
immediately within the carpel wall The diseased tissue, 
instead of being white and crisp, was a pale ashy-brown colour 
and more or less powdery. Where the disease was only partial 
a dark brown line marked the limits of the diseased portion. 
Microscopic observation showed that this zone was occupied by 
a vigorous intercalary mycelium, without haustoria, which 
brought about, probably by the secretion of a cytase, a marked 
swelling of the cell walls. ‘It had no especially marked distribu- 
tion, though, if anything, it was most abundant in the reighbour- 
hood of the vascular bundles of the cotyledons. The strongly 
thickened carpellary walls showed no signs of attack. Mixed 
with the Acrospeira I often found Penrctl/ium glaucum and 
Phoma eriophorum (Berk. and Br.). These were separated by 
means of plate cultures, and when a perfectly pure growth of 
Acrospeira was obtained, pieces of chestnut, sterilized by suc 
* It is probably a species of Haplographium. 
