646 Transactions. 
Recent work on the part of Stakman and Piemeisel (1917), and 
Stackman, Piemeisel, and Levine (1918), has shown that uredospores from 
grasses are capable of infecting cereals in the spring. For example, they 
showed (1917, p. 466) that uredospores from Agropyron, Elymus, and 
Hordeum would readily infect wheat. 
Other theories have been put forward to account for this overwintering 
of the fungus—e.g., that a perennial mycelium is present; but insufficient 
evidence was forthcoming in support, no mycelium being found in the 
host other than in the vicinity of uredosori. To account for epiphytotic 
recrudescences of wheat-rust, J. Eriksson (1897) propounded his remarkable 
mycoplasm theory, which (in brief) is to the effect that, in some manner 
not understood, the fungus is able to combine some of its protoplasm with 
that of the host, thus forming a symbiotic substance which he termed 
“mycoplasm.” This mycoplasm is supposed to remain dormant in the | 3 
leaf or seed until conditions become favourable, when the fungal elements 
become active, grow out into the leaf in the mycelial form, eventually 
giving rise to sori of uredospores! This remarkable hypothesis has been 
severely criticized by Marshall Ward (1903), who showed that the whole 
theory was based on a misinterpretation of the ordinary phenomena of 
uredospore infection. 
It was with this species that de Bary (1865, p. 25) first proved that 
heteroecism existed in the rusts. Not only is this rust heteroecious, havin 
its aecidia on one plant and its teleutospores on another, but is also known 
to consist of many biological races, each supposed to be incapable of 
infecting any host other than the one it occurs on. 
For example, Eriksson (1894, p. 292) showed by numerous cultural 
experiments that this species consisted of a number of biological races (or, 
as he termed them, “ specialisierte Formen "), as under :— 
Puccinia graminis Pers. Aecidia on Berberis vulgaris. 
l. f. Secalis—on rye. f. Airae—on Aira caespitosa. 
Tr Avenae—on oat. 5. f. Agrostidis—on Agrostis Spp- 
3. f. Tritici—on wheat. © 6. f. Poae—on Poa spp. : 
Carleton (1899), working with the same races in North America, obtained 
quite different results. He claims that there are but two biological races, 
as under :— 
l. f Tritici—on wheat, barley, cocksfoot, Agrostis alba, Festuca 
. 
B 
gigantea, Koeleria cristata, and Hordeum murinum. 
2. f. Avenae—on Avena sativa, A. fatua, A. pratensis, Hordeum 
murinum, cocksfoot, Koeleria cristata, Holcus mollis, &с. 
Jaezewski (1910, p. 321) from his cultural experiments obtained ! 
which showed interrelationships between hosts quite distinct from Eriks- —— 
son's. 0 
Again, Freeman and Johnson (1911) found that uredospores of tr nee 2 
решу infected barley, but would not infect oats, and rarely гуе; but 
Most of these results given above were obtained by using uredospores. a 
Arthur (1910, p. 228), after experimenting with teleutospores a 
€ t ot . ф E 
су gard 
