580 
Journal of Agricultural Research 
Vol. XXVI, No. is 
The percentage of entries varies from 5 to 18 in the different lots. 
It is doubtful whether these differences are of great significance. The 
question is complicated by the fact that the percentage of entries may 
vary in different parts of the same leaf. On one part of a leaf 5 out of 
73 entered, or about 7 per cent. On another part of the leaf, less than 
an inch away, 9 out of 59, or 15 per cent, entered. On one portion of 
another leaf, 5 out of 76, or 6 % per cent, entered, while on another 
portion 18 out of 133, or 13X per cent, entered. 
There evidently are local factors that influence the entry of the fungi. 
Taken as a whole, however, the evidence is fairly conclusive that so far 
as these two specialized forms of stem rust are concerned, the entry is 
not materially affected by congeniality or lack of it. The general 
average is 12 per cent of entries of form III, and 14 per cent of form XIX. 
The guard cells of Kanred stomata undergo the same change when in 
contact with appressoria that was seen in Baart. In Plate 2, B, are seen 
two empty appressoria on a stoma, one guard cell of which was broken 
in sectioning. Both fungi have entered. The substomatal vesicles on 
the inner side of the stoma (e and f) have each produced an infecting 
hypha. These have grown in opposite directions and each in turn has 
formed a vigorous-looking haustorium in an epidermal cel ] (c and d ). 
The fungus looks vigorous and thriving, the epidermal cells invaded by 
faaustoria show no disturbance, and the contents of the guard cells are 
living. The nucleus of one guard cell has shortened into a lump at a, 
but is otherwise normal in appearance. The guard cell walls facing 
the appressoria at 6, however, have lost power to stain and are glassy in 
appearance. These altered walls are not appreciably swollen; in fact, 
neither here nor in the older infections by this rust on Kanred has any 
swelling of the cell walls been observed. The walls of the epidermal cells 
adjoining an occupied stoma are sometimes slightly modified but show 
very little swelling. In this same lot of material (taken four days after 
inoculation) all gradations can be found from healthy guard cells to those 
that are dead and colorless throughout. 
Plate 2, C, is taken from material a week old. Both the appressorium 
and the guard cell are dead and disorganized and the adjoining epidermal 
cells weakened. The cell has broken from its support in sectioning. 
Stomata with guard cells killed and broken are much more common in 
this material than in Baart. An interesting point about this modifica¬ 
tion of walls exposed to the appressorium is that sometimes a small 
plate of wall material, as in the lamella in Plate 2, C, at 6, will resist the 
attack and stain normally, although surrounded on all sides by per¬ 
fectly colorless transparent walls. 
Development of Form in After Entrance 
The fungi which enter grow and develop in almost normal fashion. 
The hyphae look well nourished and have good sized nuclei (pi. 2, E, at a 
and b ), and they soon fill the intercellular spaces with a felt of inter¬ 
woven threads. Haustoria form in the usual fashion and expand fully. 
In Plate 2, D, the haustorium mother cell at b has just begun to form a 
haustorium, a, within the host call. In Plate 2, B, at c and d, are seen 
half-grown haustoria. These appear to be covered by a rather heavy 
layer of host cytoplasm but are expanding freely nevertheless; and in E 
ts a full-grown haustorium, c, connected by a narrow neck with the empty 
haustorium mother cell at d. This haustorium has extended across the 
