Howard.—On Trichosphaeria Sacchari , Massee. 629 
During the above developments the submerged hyphae 
were observed to form short branches from which chains of 
conidia, larger and darker in colour than the microconidia, 
were produced. Several likely portions of mycelium were 
fixed, and stages in the formation of these, which proved to 
be macroconidia, were observed. The short clavate hyphae 
soon showed the formation of a clear band near the apex 
which divided off the protoplasm of the globose end from 
the remainder. Five minutes later the cell-wall at the apex 
of the hypha disappeared and a spherical mass of granular 
protoplasm was extruded. A distinct cell-wall was evident 
forty minutes after, and the protoplasm was now more coarsely 
granular and showed vacuolation. After this more conidia 
were formed in basipetal succession in a chain. Their proto¬ 
plasm became vacuolated and the walls gradually darkened. 
When first extruded the protoplasm is finely granular, about 
thirty minutes afterwards several small vacuoles appear which 
gradually approach the centre and coalesce. After the forma¬ 
tion of the central vacuole the wall begins to darken, and 
in twelve hours becomes sooty-black in colour, when the 
central vacuole can no longer be observed. They measure 
on the average 22 x 15 
In many cases in this drop submerged macroconidiophores 
and collapsed aerial microconidiophores were seen to be 
developed from the same hypha, thus bearing out Massee’s 
statement that micro- and macroconidia are developed from 
the same mycelium. The drop in question was freely exposed 
to the diffused light of the laboratory, but not to direct sun¬ 
light, consequently darkness is not necessary for the formation 
of macroconidia. A similar development was observed in 
several other hanging-drops containing macroconidia only. 
Inoculation experiments were performed on healthy canes, 
using a mixture of micro- and macroconidia obtained from 
a pure culture. These were placed on a fresh wound made 
by tearing off a growing leaf, and also in wounds made in 
the cane with a sterilized needle. Neighbouring canes were 
treated in an exactly similar way, to serve as a control. In 
