146 
Journal of Agricultural Research 
Vol. XXIII, No. 3 
found. The fungus is decidedly plastic, and its individual vicissitudes 
are written in its morphology. 
These observations on the behavior of an immune host attacked by a 
parasite are not extensive enough to warrant definite conclusions, but 
they are at least suggestive. The resistance of Kanred to this strain of 
stemrust appears to depend on two and perhaps even three distinct 
characteristics—one, the nature of its stomata by which under the 
conditions of this experiment, at least, all but a few of the fungi are 
excluded, another, the formation in the invaded host cells of a substance 
or substances fatal to the fungus, and third, the heavy cell walls between 
affected and healthy host cells, by which the spread of possible toxic 
materials into healthy tissues may be prevented. 
Concerning the first point, one would suppose that a very narrow 
passage would suffice for the entrance of the fungus. Blackman (4, 
p. 339) f in describing the process of fertilization at the base of the aecium 
in Phragmidium, tells how the nucleus passes from one cell to the other 
through a very fine pore. He adds: 
Neither before nor after its passage could a pit in the wall be observed. 
Smith ( 24 ), in describing the formation of haustoria in Erysiphe, is im¬ 
pressed by the narrowness of the neck through which the nucleus and 
cytoplasm of the fungus pass to form the body of the haustorium. 
Eriksson (n) is concerned by the fact that the exit of the so-called 
mycoplasm of the cell into the intercellular spaces to form mycelium 
(or what is now believed to be the passage inward from the mycelium to 
the host cell to form the haustorium) must take place through an invisible 
opening in the wall. 
It effuses through the subtile pores that must be supposed to exist in the cell wall, 
that is to say, in the same way as the plasmodesmes between the cells. 
But, perhaps, instead of comparing the entrance of the appressorium 
to the formation of a haustorium, or to fertilization, in which the fungus 
forms a very fine pore in a wall and passes through it, one ought rather to 
compare the appressorium in its tendencies and tropisms to the hypha, 
feeling its way sensitively along the intercellular spaces and easily checked 
in its course or deflected from it by obstructions. On this basis one would 
not expect the appressorium to pass through a much narrower channel 
than a hypha would enter. 
In Kanred seedlings exposed to Puccinia graminis tritici I Pers., under 
the particular conditions of this experiment, but 1 fungus in 10 entered the 
stoma. This fact is of small importance so far as this strain of rust is 
concerned, for the immunity of this host is so great that even if all the 
fungi entered the host would not be appreciably harmed. However, 
this form of hindrance to the entrance of the fungus may not be limited 
to Kanred. Ward {29, p. 57) in his work with Puccinia dispersa Erikss. 
on bromes says: 
The vesicles and appressoria, infecting tubes, &c., may be visible much later than 
this however (i. e., four days), and it seems probable that infection can occur after 
delay, at least up to the sixth or eighth day, a point worth investigating. 
The limited work with Puccinia graminis tritici III Pers. would seem to 
show that here, too, a large majority of the fungi are excluded. Even if only 
one-fifth of the fungi entered, however, it would not always mean that 
only one-fifth as much damage would be done by the parasite. It has 
often been noted in studying inoculated seedlings of wheat and oats that 
