Infection Experiments with Various Uredinece. 185 
The results obtained were interesting enough to suggest that 
the same experiment should be tried with other rusts. Therefore 
as the season advanced and the various rusts appeared, as many as 
possible were obtained, and inoculations made with the uredospores, 
or aecidiospores as the case might be. Quite a number of the 
spores refused to germinate for reasons not apparent, so that no 
observation could be made on their infecting powers. The results 
obtained from those which did germinate are given on page 184. 
Towards the end of May Caltha palnstris was substituted for 
R. Ficcirici, as the latter plant was dying down for the summer. At 
the beginning of August it became impossible to keep Caltha alive, 
so Tropaeolum and Valeriana were used instead. 
In all these cases the germ-tube enters the stoma as in a 
normal infection on its own host plant. The after course of events 
is however quite different. The stage of development to which the 
entering tube attains before dying varies, but death ultimately 
results in all cases. In some few cases the germ-tube never gets 
right through the stoma, but forms a swelling in the actual slit and 
there stops (Fig. 1). In other cases it forms a substomatal 
swelling and emits from this one or more hyphae (sometimes as 
many as six, Fig. 2) just as in normal infection. These hyphae 
may grow to a greater or less length before dying, from a mere 
projection to three-fourths the depth of the spongy parenchyma 
(Figs. 3, 4, 5, 6, 7 and 8). A very common case is that in which 
the germ-tube forms, in the stomatal slit, a swelling which projects 
into the substomatal space. This swelling may develope no farther 
(Figs. 9 and 10) or from the side of it a hypha may grow till it 
meets with a cell and at the point of contact form a swelling and so 
ends its growth (Figs. 11, 12, 13 and 27). Another very common 
case is for the germ-tube to form no swelling at all, either in or 
under the stoma, but to grow straight on till it meets a cell (which 
may be at a considerable depth in the leaf) and there form a 
swelling (Figs. 14, 15 and 16) or else swell along its whole length 
and so come to an end (Figs. 17 and 18), though in rare cases the 
swelling emits a hypha (Fig. 19). 
From this it will be seen that in the struggle against adverse 
circumstances the entering germ-tube tries all kinds of plans quite 
different from its behaviour in normal infection. 
The length of time that these hyphae remain capable of growth 
varies ; but in no leaves four days after inoculation can hyphae be 
found which have not obviously reached the end of their capacity 
for growth. In some cases in two-day preparations, hyphae are 
