586 
Journal of Agricultural Research 
Vol. XXVI, No. 12 
part of the hypha being in one section, C v and the distal part (separated 
by a space in the drawing) in the next section, C 2 . The haustorium 
mother cell, with its pair of minute nuclei, is at a, and the young hausto¬ 
rium forming from it at 6. There is no marked disturbance of the host 
cell contents, but the plastids and nucleus of the host cell are beginning 
to collect around the haustorium. 
This is the beginning of a violent reaction in the host cell. A slightly 
later stage, also taken from two-day material, is shown in Plate 4, D. 
There has been a rapid flow of the host cell contents toward the hausto¬ 
rium at a. Two lobes of the host cell at c and d have been completely 
evacuated and their walls have collapsed. Cytoplasm, plastids, and 
nucleus b are massed around the haustorium. The cytoplasm and plastids 
seem to be still alive, but the chromatin network of the nucleus is nearly 
dissolved. The fungus is less violently affected, although the haustorium 
mother cell at e has collapsed and is dying. In this case, the fungus has 
strength for a second attack, as there is a fresh young hypha at /. 
A more advanced condition (still taken from two-day material) is 
represented in Plate 5, A. The infecting hypha and haustorium mother 
cell at the end of the guard cell at a are shriveled and dead. The attacked 
mesophyll cell also is dead. It is stained a deep red at the end near the 
fungus, and the color fades toward the other end. Evidently here, too, 
there was a concentration of living matter about the fungus. The 
haustorium b lies within a narrow clear zone, making it a conspicuous 
object within the cell. This clear space was present, but less sharply 
defined, in the earlier stages. The host nucleus c, which is pressed close 
against the haustorium, is dead and has lost all trace of inner structure. 
The damage in this case does not stop with the cell directly attacked. 
At d , where the dead cell touches another cell, we find the living cell 
slightly plasmolyzed, its nearest cytoplasm altered in appearance and 
the contact wall between the two cells slightly swollen. At e , a contact 
with a third cell at a greater distance from the fungus, the damage is 
negligible. One or two plastids have disintegrated, but there are no 
other visible changes. 
When the first haustorium is formed in a mesophyll cell, it usually 
results in the massing of a large part of the cell contents about the 
haustorium, followed by the immediate death of both the cell and the 
haustorium. This is a severe check on the fungus, as its limited resources 
are seriously depleted. When the fungus possesses vigor enough for a 
second attack, it goes on. In looking through the older material, however, 
it is not unusual to find minute infections consisting of an empty appres- 
sorium and substomatal vesicle, dead colorless guard cells, and a single 
dead mesophyll cell. This shows that the fungus may die after making a 
single haustorium. More commonly, however, several host cells are 
attacked in succession before the fungus is exhausted. For a time these 
later attacks result like the first. 
Still later, however, the host reacts less violently to the fungus. The 
milder reaction may be due to enfeeblement of the fungus; or to some 
response in the adjacent host tissues to the presence of the fungus; or, 
perhaps, to the fact that the fungi capable of evoking the most violent 
reaction in the host are already killed by it; or, conceivably, even to a 
slightly varying resistance to the rust in different parts of the same leaf. 
The latter is least likely, as it would be difficult to explain why it is 
invariably the first cells attacked that respond most violently. 
