CERTAIN BRITISH HETERCECISMAL UREDINES. 
] 63 
R. auricomus (252, 273), but with no result. Hence it 
appears that Uromycesdactylidis has its secid iospores upon 
R. bulbosus only, and nut upon R. repens or acris. 
Spores from the iEcidia on R. bulbosus were placed upon 
Dactylis glomerata (279, 289, 298), where they in all cases 
gave rise to a Uredo followed by the Uromyces dactylidis. 
iEcidiospores of this iEcidium placed upon Poa pratensis (299) 
and P. amma (300) gave no result. These last two cultures 
were duplicated with expt. 298. 
In no case, however, in which the Uredo was produced upon 
Dactylis could I find the least trace of any paraphyses, nor 
could any be found upon the Dactylis uredo, as it occurs 
naturally here. The question, therefore, presents itself, Have 
Dr. Schroter and myself the same fungus in view ? It is very 
unlikely that there should be two Uromyces upon Dactylis 
both having their iEcidia upon R. repens. Personally, I 
rather incline to the belief, and it is only a belief which subse- 
quent observation must confirm or disprove, that these 
paraphyses are found in certain conditions of the Uredo and 
not in others; that in other words, their value as a specific 
character is not of vital importance. But what these condi- 
tions are which favour the development of paraphyses, I am 
unable, at present at any rate, to say. Just as I was unable to 
discover any difference externally between the two iEcidia 
upon R. repens, so am I unable to point out any anatomical 
differences between the two iEcidia on R. bulbosus, 
namely, that of Uromyces dactylidis and Puccinia 
Magnusiana. Physiologically, however, they are distinct 
enough. 
