104 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
of the method of spore delimitation in Saprolegnia. While there 
is but one layer of spores in this species, there are three in P. macro - 
sporus , and in both there is no epiplasm. It is to be regretted that 
Popta examined living material only. 
And finally Juel (:02) has described the process of sporogenesis 
in Taphriclium cdgeriense. According to his description the pro¬ 
toplasm is differentiated into a denser portion that lies at the peri¬ 
phery and in which the nuclei are arranged in a single layer, and a 
vacuolated portion that fills the rest of the sporangium. Uninu- 
cleated spores are now formed from the dense protoplasm, but Juel 
could not determine how delimitation was effected. 
The details of spore formation in the Hemiasci given us by these 
investigators are apparently too meager to enable us to come to a 
final conclusion, but it seems quite unlikely from such as are known 
that the “ sporangia ” of the Hemiasci are intermediate between 
Mortierella-like sporangia and asci. It is evident that the latter are 
not only not directly connected, but also that they are not linked to 
one another by means of such forms as Protomyces and Taphridium. 
Nevertheless I cannot agree with Harper ('99) that “the total dis¬ 
similarity of the process of cleavage in the sporangia described 
[Pilobolus and Sporodinia] and the ascus, makes it necessary to look 
for the ancestors of the Ascomvcetes elsewhere than in the lower 
Fungi,” for it is to be noted that the forms he investigated were of 
a highly specialized type. If the ancestral homologue of the ascus 
is a sporangium at all, it is more likely to be found among more 
primitive sporangia than Harper described. This brings us to the 
second suggestion, namely, that the ascus is a derivative of an 
Oomycetous zoosporangium. 
Spore formation in the lower Phycomycetes has received some 
attention at the hands of Ilartog ('88), Rothert, Humphrey (’93) and 
more recently Davis (:03). In their descriptions of the cytology of 
the zoosporangia of the Saprolegniaceae they agree on the general 
features. The central part of the sporangium becomes vacuolar, 
the vacuoles flow together and form a large cavity filled with sap. 
The protoplasm thus occupies a peripheral position. Radial clefts 
then start from the vacuolar membrane cutting the cylinder of pro¬ 
toplasm into the spore origins, which contain a single nucleus each, 
and which Avithout further division are transformed into zoospores. 
Leaving aside for a moment their relative ontogenetic values it 
