478 BOTANICAL GAZETTE [DECEMBER 
centers of attraction, about which the protoplasmic lining of the walls 
is collected. ‘The whole process is a form of free cell formation, in 
which the entire protoplasm is utilized, without involving the forma- 
tion of a periplasm. He holds, therefore, that the sporangium of 
the Saprolegniaceae represents an advance over forms in which peri- 
plasm is formed. The sporangium is differentiated into central 
vacuole and peripheral protoplasm, and is perhaps a stratified struc- 
ture in itself, whose polarity is determined by the position of the nuclei, 
which in turn influence the position of the spores, as has been pointed 
out by HARPER (’99). 
According to the work of RorHErt (’88), Hartoc (’88), Hum- 
PHREY (’92), TRow (’95), and Davis (:03), the sporangium is multi- 
nucleate when cut off, the nuclei lying scattered in the peripheral 
layer of protoplasm. Davis practically confirms the account of the 
earlier authors. The uninucleate spores originate by means of clefts, 
which proceed from the central vacuole of the sporangium to the per- 
iphery, dividing the protoplasm into polygonal areas. The spores 
are later formed from these uninucleate areas. There is no mitotic 
division of the nuclei or cytoplasmic centers (coenocentra) in the 
zoosporangia. 
HUMPHREY (’92) first studied oogenesis in Saprolegnia by means 
of modern technique, but was soon followed by Trow (’95, ’99) and 
Hartoec (95, 96, 99). There seems to be a great diversity of opin- 
ion as to the behavior of the nuclei, which far exceed the ultimate 
number used in the formation of eggs. Humpurey and Harroc 
believe that the nuclei fuse in groups to form the functional nuclei. 
Trow claims that many nuclei degenerate until the requisite number 
is reached, which results Davis (:03) has confirmed, but differs 
from TRow in regard to the sexuality of the Saprolegniaceae. The 
oogonium arises as an enlargement of the end of a hypha, into which 
passes a dense mass of cytoplasm and nuclei. A central vacuole is 
formed, with a peripheral layer of protoplasm lining the walls in 
which lie the nuclei. The nuclei divide once by mitosis. ‘The proto- 
plasm aggregates into masses which form the eggs. The process of 
separating these masses by means of a series of fusing vacuoles, has 
been described by Davis. He finds that the egg initials are formed 
about cytoplasmic centers (coenocentra), much as has been described 
