confluens and the., Morphology of the Ascocarp. 391 
show no evidence of their existence till they burst through 
the enveloping medium in an advanced stage of development. 
When we further consider that the coenocytic condition is 
very common, and that the function of the cells can only be 
determined through a knowledge of the behaviour of their 
nuclei, which are frequently minute in size, it is easy to see 
why our knowledge has not progressed more rapidly. 
In this state of affairs it is certainly scientific as far as 
possible to withhold judgement as to questions of relationship, 
and above all to refrain from the attempt to force little-known 
forms together into artificial groups based on mere superficial 
resemblances. For the present the question as to the relation 
of Pyronema to the Erysipheae should be regarded as awaiting 
for its solution a further knowledge of such forms at least as 
are most closely related to these types, the remainder of the 
Perisporieae on the one hand and the simpler Discomycetes 
on the other. It is, however, sufficiently evident that the 
differences between Pyronema , Sphaerotheca , and the Lichens, 
cannot be considered as affording sufficient reason for regarding 
the Ascomycetes as a polyphyletic group. The ascus with 
its peculiar method of free cell-formation which has not been 
observed elsewhere as yet among Algae or Fungi is strong 
evidence for the morphological unity of the group. 
Assuming that the ascocarp is always as it has been shown 
to be in the cases cited, the asexual spore-fruit, developed 
from a fertilized or possibly in some cases a parthenogenetic 
egg and its envelopes, the question as to the nature and 
relationships of the group of the Exoasci seems a difficult one. 
It would be very interesting to know, in this connexion, how 
nearly the method of spore-formation in this group corresponds 
to the method of free cell-formation as I have found it in the 
Erysipheae and Discomycetes. I have myself begun some 
studies looking to this end, which are not yet concluded. 
As to the morphology and relationships of the group as 
a whole it is known that Brefeld regards them as very simple 
forms in which the asci have not come to be enveloped in 
more or less closed envelopes. Dangeard (7) describes the 
