84 BOTANICAL GAZETTE [AUGUST 
formation of sporangia in Pilobolus and Sporodinia, and by Dean 
SWINGLE (’03) in Rhizopus and Phycomyces. The plasma membranes 
are formed chiefly along the surfaces of flattened vacuoles. I have 
seen no evidence of furrows cutting inward from the surface as takes 
place in Pilobolus and Rhizopus, but such structures would not be 
seen easily in Vaucheria because the stalk is rather narrow. Fig. 4 ~ 
illustrates an arrangement of vacuoles at the base of an oogonium 
which would probably have determined the position of the cross 
wall, and fig. 5 shows a more advanced condition when the plasma 
membranes have definitely separated from one another. The stages 
shown in figs. 6-8 are all somewhat older than that of fig. 5, and 
there is evidence in all of them that the wall has began to form as a 
delicate film visible at certain points between the two plasma mem- 
branes. 
The oogonium at the time of the formation of the cross wall is 
multinucleate. Thcre is no evidence in my material of the with 
drawal of nuclei from this structure, as described by OLTMANNS, oe 
before the wall is laid down, and my preparations indicate that the _ 
oogonium contains as many nuclei immediately after the formation 
of the wall as before. These conditions are shown in figs. 5-7: 
The number of nuclei is always large, but is variable, ranging from 
20 to 50. They may be readily demonstrated at this period when 
properly stained, but are very difficult to trace from this stage of 
development onward because of the remarkable and rapid nuclear | 
degeneration which sets in at this time. This interesting phenomenon 
is apparently exactly the same as that which takes place at closely 
corresponding periods of oogenesis in Saprolegnia and in several of 
the Peronosporales. 
The degeneration of the nuclei really begins a little before the 
oogonium is separated from its parent filament. At that time the 
nuclei in the oogonium do not stain as strongly as those in the anthe 
ridium and in neighboring portions of thé vegetative filament. The 
nuclear membrane is less distinct and there is very little substance 
in the nucleus except the large nucleolus. Fig. g presents a series 
nuclei from the same section of which fig. 7 is a single oogonium. 
n fig. 9, a is a nucleus from the antheridium, 6 from a region of es 
branch slightly below the antheridium, ¢ from near the oogonlums 
hd 
