232 DIVISION II.—-COURSE OF DEVELOPMENT OF FUNGI. 
in special receptacles, as adaptations to special circumstances of development, the 
further consequence of which is the establishment of diclinism. 
It must be admitted that comparisons of this kind are always somewhat insecure 
and readily afford too much room for the play of fancy, if they are not supported by 
a more complete series of well-known intermediate forms than can at present be 
produced in this case. The comparison in the text may be offered with this 
reservation attached to it. It appears to me to fit in quite naturally with ascertained 
facts; nor is it too much to expect that the intermediate forms that are wanted will 
be found, if we consider how few of the whole number of Ascomycetes have been 
examined up to the present time. 
If we compare the whole course of development of the Ascomycetes in which 
this course is fully known with that of the groups of Fungi described in previous 
sections, we become distinctly aware of a parallelism between Eremascus and the 
Ascomycetes, which are provided with archicarps and antheridial branch on the 
one side, and the Mucorini, Peronosporeae, and Saprolegnieae on the other. A 
thallus proceeds from the carpospores (ascospores, oospores), and terminates its 
development with the formation of an archicarp and antheridial branch and of 
the new carpospore which is formed from them. To this the whole course of the 
development is limited in many cases, as in Eremascus, Pyronema, and species 
of Ascobolus on the one side, and on the other in Pythium vexans and Artotrogus; 
in most cases it is interrupted by the formation of other spores, the gonidia. The 
gonidia of a species are sometimes all of one kind, as in Erysiphe and Peronospora, 
sometimes of more than one kind in the same species. The parallelism extends 
even to close resemblance in form in organs of the same name in certain groups. 
Eremascus might, to judge from Eidam’s description, be almost ranked with the 
Mucorini, especially the Piptocephalideae ; on the other hand, none of the essential 
developmental characters of an Ascomycete are wanting in it. In the form of its 
archicarps it is perfectly like Penicillium, Gymnoascus, Eurotium and other genera. 
There is also a great amount of agreement in respect of the thallus, formation of 
the gonidia, archicarp, and antheridial branch between the Erysipheae, especially 
Podosphaera, on the one hand, and some Peronosporeae on the other. These 
groups therefore establish a nearer connection between the Ascomycetes in question 
and the Peronosporeae; a convergence of the two groups which amounts to 
actual contact and may be regarded as phylogenetic affinity. This connection with 
the Peronosporeae is closest in Podosphaera, because there caeferis paribus not 
only the archicarp and antheridial branch are like the same parts in the Pero- 
nosporeae, as, for example, in Phytophthora omnivora, but the ultimate develop- 
ment also of the archicarp only goes a little further; the eight-spored ascus with 
its stalk proceeds from division of the cell twice repeated, while in Phytophthora the 
archicarp becomes the oogonium with the oospore. Erysiphe is closely con- 
nected with Podosphaera, in which the archicarp by repeated cell-division and 
branching gives rise to a number of asci, and all the rest of the Ascomycetes in 
question with Erysiphe, as appears plainly from the details which have been given 

! For a more extended comparison see Beitr. IV, p. 109. 
