482 BOTANICAL GAZETTE [DECEMBER 
how the ascocarps of Galactinia, Acetabulum, and Pustularia arise, 
and whether apogamy or parthenogenesis is associated with the appear- 
ance of binucleate cells in the ascogenous hyphae. 
The cells of the ascogenous hyphae of Thecotheus are not binu- 
cleate, and I am inclined to accept for this form HARPER’s interpre- 
tation that the asci are spore mother cells, modified by adaptation 
as explosive organs and as reservoirs of reserve food supply, in which 
a merely vegetative fusion has occurred to maintain a definite nucleo- 
cytoplasmic relationship. A triple division follows to complete the 
reduction and distribution of the somatic chromosomes to each of 
the resulting eight nuclei. The sporophyte would thus include the 
ascogenous hyphae and the asci up to the time of the reduction 
division, which initiates the gametophyte generation. 
In the typical ascus the nuclei of the eight spores contain the gam- 
etophyte number of chromosomes, as would. also be true when only 
two or four spores are formed. When any of these spores germinate, 
they give rise to true gametophyte structures, usually a septate 
mycelium, which may reproduce itself asexually by conidia before 
sexual reproduction, as in Eurotium or Erysiphe. It is well known 
that many ascospores contain more than one nucleus, and, as FAULL 
and others have shown, these nuclei are formed by mitotic division 
of the primary spore nucleus. The spore may be septate or non- 
septate. The typical ascospore is uninucleate and non-septate. 
Both the number of nuclei and the number of septa in a spore vary 
from one to many. Spores which are septate have apparently begun 
an intrasporal germination, the gametophyte forming considerable 
embryonic tissue within the old spore wall. Spores which are not 
septate but multinucleate have also undergone embryonic develop- 
ment, but without cell division. 
We perhaps should expect from what we know of other Ascomy- 
cetes that in many-spored asci, as in Thecotheus and Ryparobius, 
spores would be delimited as soon as eight free nuclei were formed. 
This does not occur in either of these forms, but further nuclear divi- 
sions take place before the spores are delimited. A closer analysis, 
however, shows that we have analogies for these conditions in the 
behavior of other undoubted spore mother cells. Intrasporal germ- 
ination may be looked upon as comparable to that which occurs in the 
