ORIGIN OF THE BASIDIOMYCETES. 441 
with any higher form of fruit, and in many instances it would 
appear that such must be considered as entities or species; the 
facts in favour of such an argument being the power to repro- 
duee themselves apparently indefinitely, and the absence of 
proof as to the existence of any other phase of reproduction in 
their life-cycle. 
Assuming this statement to prove correct, it suggests the 
following problem :— 
How long must a conidial form continue to reproduce itself 
after the disappearance of its higher phase in the cycle of 
development, before it can be cousidered as a species in the 
ordinary acceptance of that term ? 
If taken to task as to the evidence of a second phase having 
existed at any previous period in the life-history of such 
organisms, it may be stated that presumable evidence is forth- 
coming, not only in the genera Stilbum and Tubercularia, but in 
numerous other instances where a conidial or simple phase has, 
from analogy, lost for all time the higher stage of fruit it once 
possessed. 
The evidence obtainable in the genera under consideration 
consists in the presence, in several iustauces, of a more or less 
well-developed stroma, which, however, only produces conidia, 
whereas, following the sequence of gradual disappearance of the 
stroma, we come to species where this primordial strueture has 
entirely disappeared. 
In the form-genus Botrytis, we have a similar decadence of the 
stroma or sclerotium, whieh in some species produces the asci- 
gerous condition only, in others the ascigerous or conidial 
eondition, depending on external conditions, in others again 
giving origin to conidia only; whereas in a host of other species 
the stroma is quite rudimentary or entirely absent, the conidial 
condition alone remaining. 
It is in the Uredinez, however, that we encounter the clearest 
evidence of the gradual disappearance of one or other of the 
forms originally included in the life-cycle of the various species. 
After the announcement of Juel's discovery, I examined nume- 
rous species of Stilbum and Tubercularia, for the purpose of 
ascertaining whether some other members might not prove on 
careful examination also to belong to the Protobasidiomycetes. 
Among those examined was the S£ilbum-condition of Sphero- 
stilbe microspora, Cooke & Massee ; and here I was much surprised 
