576 
Journal of Agricultural Research v 0 i. xxvi, no. « 
centers of infection, which result ultimately in a circle of secondary 
uredinia around the first. Similar stolons occur in P. glumarum (Schm.) 
Erikss. and Henn., as described by Pole Evans ( 12 , p. 452 ), and in 
P. dispersa Erikss., as described by Ward ( 44 , p. 37). 
The greater part of this surplus food, however, is utilized in spore 
formation. By the seventh day, at the center of the infection, hyphae 
are massing under the epidermis antecedent to the formation of uredinia. 
In Plate 1, F, is shown a portion of a young uredinium, drawn on a 
smaller scale than the preceding figures. It happens here that the uredin¬ 
ium is forming under the same stoma through which the fungus entered. 
The old collapsed wall of the appressorium is still visible at a. In this 
case the guard-cell wall was but slightly affected by the fungus, there 
being a small pale spot on the inner wall below the appressorium. There 
is a felt of hyphae filling the intercellular spaces and running up into 
the substomatal chamber where the young binucleate spores are form¬ 
ing. The hyphae at the center of the infection and even up near the 
uredinium are well drained of their contents; in fact, are nearly empty, 
and the food is now concentrated in the spores and the layer of hyphae 
just below them. 
As growth proceeds, the young spores lift the epidermis, rupture it, 
and both the spores and their stalks elongate. In Plate 1, G, is represented 
a narrow strip through the center of an infection two weeks old. Spore 
formation is taking place on both surfaces of the leaf, a and b. There 
is a massive development of mycelium just below the surface, and both 
here and in the central region the mycelium is devoid of contents. 
Numerous spores have already been freed, as can be determined from the 
withered stalks that remain. A few fresh spores are still being formed. 
Some of the host cells are crowded out of shape and almost obliterated, 
but in many cases are still living. 
In older infections there were noted for the first time a few scattered 
haustoria which had the appearance of having formed in an uncongenial 
environment. They were only partly expanded and were surrounded 
by a thick, glistening, transparent covering. It may be that they indi¬ 
cate some feeble and tardily developing resistance to the rust which 
shows in but a small percentage of the host cells. The rest of the haus¬ 
toria looked normal even when the contents of the host cell had collapsed. 
Effect of Neuclei and Plastids of Baart 
This luxuriant growth of the fungus takes place with the minimum 
of harm to the host, but a close comparison of successive stages shows 
that the host tissues within range of the fungus undergo a fairly definite 
series of metabolic changes. The most tangible evidence of this is the 
alteration in size and shape of the host nuclei and plastids. 
The nuclei in the affected area expand rapidly, remain large for several 
days, and then collapse and die. Typical stages in this process are re¬ 
presented in Plate 2, A, a to h. The same magnification (X 1130) was 
used throughout the series. The normal nucleus, a, uninfluenced by 
the fungus, has a heavy membrane, a relatively small nucleolus and a 
nuclear net which is usually very dense but occasionally shows open 
places. When the fungus is six or seven days old, nuclei of the mesophyll 
cells in the center of the infection begin to expand. The nuclear net 
opens irregularly, forming a larger mesh along one side (PI. 2, A, b and 
c), as if the nucleus were taking in water on this side. Nucleolar material 
