[Vor. 3 
46 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
ored green. Longitudinal sections showed the 
hyphae of the fungus running longitudinally in 
the lumina of the spiral vessels and the bast 
fibres. Cross-sections showed only the cut ends 
of the fungus. Drawings illustrating these facts 
are shown in figs. 8 and 9 which were made 
with the aid of a camera lucida. It was found 
that besides the purely vegetative hyphae, the 
fungus produced conidia in the vessels of the 
host (fig. 10). Those spores observed in the 
host tissue were all of the unicellular type. 
All the evidence shows that the fungus at- 
tacks the root first, but Just how remains to be 
worked out. After entering the host, it is con- 
fined to the vascular system. The fungus was 
never isolated from 
42-2 the stem until gcn Ñ 
tudinal section of a marked yellowing 
in diseased of the leaves ap- 
Mr уа, peared, although it 
hae ZZ Was always present 
Camera lucida in the tissues before 
sketch X 1000. they had been killed. 
This fact is brought out in pl. 2, 
fig. 12 in which is shown a branched 
plant, one branch of which was at- 
tacked, while the other remained 
healthy in appearance. The leaf at E 
was still alive although one side of it 
was yellow. The stem of the plant 
appeared normal externally. The 
fungus, however, was isolated from the 
stem below the branching and Fig. 10. Longitudinal section of 
ALI points D qud Z pu thie Ме Geet ote инш, 
diseased branch, while the parts Stained with Pianeze IIIb. Camera 
at B and C on the other branch BEER 
remained sterile. The plate made from this plant is shown 
in pl. 2, fig. 13. 
