CERTAIN BRITISH HETERCECISMAL UREDINES. 
155 
R. repens gave no result (118, 249, 254, 270, 290 ) ; we may 
therefore conclude that Uromyces dactylidis is not con- 
nected with the iEcidium upon R. repens. 
On the other hand, the iEcidium on R. repens was found 
to produce the Uredo and Uromyces uponPoa trivialis (190, 
322, 333). But here another difficulty confronted me, inas- 
much as no less than in seven other cultures with the spores 
of the iEcidium on R. repens applied to P. trivialis and 
P. pratensis did I fail to obtain any result (120, 146, 147, 
155, 191, 336, 370). It is all very well to say that one 
positive result is of more value than an indefinite number of 
negative results, but succeeding only in three out of ten cultures 
requires some satisfactory explanation. Those who have per- 
formed cultures with the Uredines know well enough how easy 
it is to fail from a variety of causes. I have failed more than 
once in infected wheat plants with Uredo linearis in which 
no question of specific identity is concerned. It is further- 
more very easy to fail with cultures in which secidiospores are 
employed as the infecting material, because, in the first place, 
the secidial cup is full of spores, but only the few mature ones 
at its orifice will germinate at all ; and in the second place, 
because even these ripe spores very rapidly lose their germi- 
native power. Still these facts were well enough known to 
me, at any rate in my later cultures when I had also gained 
some knowledge of the minutiae required for successfully 
manipulating with these bodies. As will be seen later, the 
explanation simply is that upon Ranunculus repens 
another iEcidium occurs (that of P. Magnusiana) which so 
closely resembles it in appearance, and in the form, size, and 
colour of its spores, that I am quite unable to tell the one from 
the other. I do not say that this cannot be done by others 
more skilled in the differentiation of uredine spore forms, but 
up to the present I have been unable to do so. 
The iEcidium on R. ficaria gave no result on P. nemoralis 
(133, 296), nor upon Dactylis glomerata (297), but upon 
P. trivialis (295) it gave rise to the Uredo of Uromyces 
Poae. 
