a Fungus causing a Disease of the Sugar-cane . 5 2 1 
twelve days there was no sign of any disturbance of the 
tissues due to the entrance of a parasite, although some of 
the spores had germinated on the surface. Repeated experi- 
ments with conidia on an unbroken surface of young stems 
and leaves also gave negative results ; hence it may be con- 
cluded that conidia of the Melanconium phase can only obtain 
an entrance into the cane through a broken surface, and 
judging from an examination of diseased canes, the most 
vulnerable points are the bases of dead leaves low down on 
the cane. The bases of broken-off lateral shoots and rootlets, 
and the burrows of the ‘ shot-borer ’ also favour the entrance 
of the fungus. 
When the vegetative hyphae of the Melanconium have 
become abundant in the tissues of the sugar-cane, they form 
dense wefts immediately below the epidermis and between 
two bundles ; the weft commences to grow in a very compact 
manner and eventually forms a dark-coloured parenchymatous 
cushion or stroma composed of small polygonal cells about 
5 jot in diameter ; from 1-3 loculi are present in the mature 
stroma ; the cells lining the loculi each send out one delicate 
spine-like hyaline sterigma or pedicel which in turn bears 
a narrowly cylindrical, uniseptate, pale brown, straight or 
very slightly curved conidium measuring 14-15x3*5 — 4 /x 
(Fig. 4). When mature the stromata vary from 1-2 mm. 
in length, and owing to development taking place in the 
direction of the epidermis, the latter eventually splits and 
the black surface of the stroma protrudes through the crack. 
When mature the conidia ooze out of the cavity in which they 
were formed through a minute pore or ostiolum at its apex, 
and being held together by mucus derived from the deli- 
quescence of the external portion of the exospore, remain in 
the form of minute black, more or less curled filaments, to 
2-3 mm. long which become rigid when exposed to the air 
(Figs. 2 and 3). 
The Melanconium form of fruit was not reproduced in any 
of the flask-cultures containing the nutrient solution only, but 
the characteristic stroma and conidia — both colourless — were 
