550 
PROFESSOR H. MARSHALL WARD OH THE 
whereas some forms, such as Ustilago destruens first develope a definite mycelium 
from the resting-spore, and the yeast-like cells are budded off from this into the air, 
in other cases the budding commences at once on the germination of the spore. Thus, 
in Ustilago antlierarum ,+ the budding begins direct from the spore, or immediately the 
promycelium is protruded. Similarly with Ustilago intermedia ,j U. Maydis, and others. 
Now, if we supposed the spore to germinate in this manner while still attached to the 
mycelium inside the plant, we should have something very like wffiat happens in the 
Fungus I have described. In fact, I regard the swollen parts of the mycelium which 
give rise to the gemmules as the liomologues of the spores proper {i.e., resting-spores) 
in the ordinary Ustilaginese. Anyone who compares the development of the spores 
in those Ustilaginese which have been sufficiently studied § cannot fail to notice the 
remarkable similarity to the development of the branches which bud off the gemmules 
in this Fungus. If, for any reason, the future resting-spore of an Ustilago began to 
put forth buds (gemmules) before it developed its thick coats, we should have the 
very process I have been describing. In the case of the Fungus causing the tubercles 
on the roots of Leguminosse it is not difficult to suggest that it might obviously be a 
disadvantage to the parasite to develope resting-spores, of (comparatively) large size, 
which should be set free on the rotting of the root, since they would not easily be 
disseminated in the soil; on the other hand, it would seem to be a great step gained 
if the habit of forming yeast-like gemmules no larger than Bacteria was thrown back, 
as it were, earlier into the life-history. There is no lack of equally curious adaptations 
in the life-history of Fungi. In fact, steps towards such retrogression are already 
found in the asexual development of the resting-spores of the Ustilaginese if we accept 
de Bary’s views as to their homologies; and it does not seem possible to refute them. 
These extremely minute gemmules will be disseminated between the interstices of 
the soil as easily as the well-known Schizomycetes found everywhere ; every stream, 
every puff of wind, and many slighter movements will carry them from place to place, 
and hence their ubiquity. 
I have been much exercised with the question as to whether these gemmules bud 
and multiply in the soil, i.e., in artificial nutritive solutions, and have repeatedly 
made efforts to cultivate them in hanging drops of pure water, Pasteur’s solution, and 
other nutritive media. The general result has been disappointment. In some cases 
I have thought that certain dense flocculent groups which develope in Pasteur’s 
solution were clouds of the buds, but they never form rapidly, and in the interval (four 
to eight days, or longer) Bacteria have always made their appearance. The suspicion 
that these flocculent clouds might be colonies of budding gemmules is not an absurd 
one : very similar, I believe identical, clouds form on the root-hairs of my water- 
* Loc. civ., Plate VII., fig. 24. 
f Loc. cit., Plate I., especially figs. 14-17. 
J Loc. cit., Plate VI. and Plate IV. 
§ E.g., de Bary, ‘ Morpliol. nnd Biol.,’ fig. 82, p. 189 (Engl. Edition, p. 175). 
