127 
The foregoing account of Puccinia graminis includes the principal facts at 
present known concerning its structure and life-history. That one stage of its 
existence is passed as a parasite upon the barberry does not admit of doubt. 
But the question presents itself—is this an absolutely essential stage, or can the 
mildew be propagated without the intervention of the eidiwm state? The 
principal reason which suggests the question is the disproportion which exists in 
England between the amount of mildew aud the number of barberry bushes. 
Especially does this obtain in Norfolk, where this shrub has been, to a great ex- 
tent, exterminated. It isa matter of common observation that stray wheat plants 
grown on a heap of farm-yard manure, are almost invariably attacked by mildew. 
This has been asserted to be due to over-nitrogenous stimulation, but no amount 
of nitrogen has ever yet produced a Uredo or any other spore. It is worth re- 
membering that a wheat plant, so situate, must, during the spring, be surrounded 
by an atmosphere loaded with promycelium spores from the Puccinia, which is 
invariably formed upon straw, and it is only reasonable to suppose that these 
spores have a greater chance of infecting a plant than when in artificial culture a 
limited number is placed before the fragment of a leaf, and examined under the 
microscope. 
But, assuming that De Bary is right in his view, that these promycelium- 
spores cannot enter the wheat plant, is there any way in which the Uredo can be 
produced other than by the implantation of Aicidium-spores? It is improbable 
that the mycelium is perennial, even when affecting perennial grasses.* During 
the winter months, in a not exceptionally severe winter, it is possible to find stray 
pustules of Uredo either upon wheat, or, more commonly, upon Twitch, growing 
in sheltered situations. Thus, last December (1881), I found a fresh pustule of 
Uredo upon Twitch (7. repens) in a wood, and also one in the month of March on 
a hedge-bank at Flitcham. This spring, our Norfolk and Suffolk wheats were 
much affected with rust; some of this may be, and probably was, due to Uredo 
linearis, kept alive from the previous autumn, but the bulk of it was due to the 
Uredo of Puccinia straminis, which is always an earlier Uredo than that of 
P. graminis. 
There is a wonderful difference in the amount of injury done by mildew, 
when derived directly from the barberry, and when derived from Uredo 
that has reproduced itself through several generations. This is very obvious 
from the fact that the Uredo is to be found every year in almost every—if not in 
every—cornfield, but the farmer takes no notice of it, as it does not appreciably 
diminish the yield. But with the mildew which occurs in the middle of the bar- 
berry bush the case is different. The fungus grows with such energy that it so 
injures the wheat plant as to prevent it producing more than a few starved ker- 
nels. With such vigour does the mycelium grow and fructify at the expense of 
the wheat, that the straw of the latter frequently does not ripen, but dies green. 
This is only what one would expect when the fact is taken into consideration, that 
the Aicidium-spore is a sexual product, whereas the Uredo-spore is not. 
* De Bary, oc. cit., p. 23. 
