470 Yasui. — On the Life-history of Salvinia natans. 
In 1887 Campbell, in a paper dealing with the development of plant- 
spermatozoids, described the development of the male gametophyte of 
Salvinia . He agrees with Pringsheim (’ 63 ), who stated that there is one 
antheridium containing eight spermatozoids in each male prothallium. 
Belajeff (’ 98 ) made a careful study of the development of the male pro- 
thallium of Salvinia. He mentions that there are two antheridia in a single 
male prothallium, differing from the statements of Pringsheim and Campbell. 
Campbell still maintains his older view in the second edition of ‘ Mosses and 
Ferns ’ (1905). 
The material for the study of spore-formation was collected in October, 
and the gametogenesis, fertilization, and embryogeny were studied from the 
material raised from spores in the laboratory from January to March during 
the last two years. The material was fixed either with Flemming’s weaker 
solution or with chromo-acetic acid mixture. It was embedded in paraffin 
with the melting-point 52 0 C. The studies were made almost entirely with 
microtome sections. 
Microspore Formation. 
The microsporangium-initials are already formed on the surface of 
the very young sporocarp, as shown in the Text-figure. Each initial-cell 
undergoes a transverse division by which the stalk-cell is cut off from the 
apical cell (PL XLI, Figs. 1, 2). The stalk-cell then divides two or three 
times transversely. The apical cell 
segments again, cutting off the basal 
cell, the first primary wall-cell, below 
(Fig. 3, w-i). The apical cell then 
divides by a wall more or less perpen- 
dicular to that of the basal cell (Fig. 4, 
w 2 ). It is soon followed by a peri- 
clinal wall (Fig. 6 , w 3 ), thus cutting 
off three primary wall-cells and the 
archesporial cell inside. According to 
Juranyi (’ 73 ), who made the first care- 
ful observations on the development 
of both the micro- and macrosporan- 
gium, one more wall-cell is cut off, 
thus forming four primary wall-cells 
enclosing the archesporial cell instead of three as I have observed (compare 
Juranyi’s PI. I, Figs. 6 , 7). 
The first-formed cell of two upper primary wall-cells then divides 
by a wall parallel to the long axis of the sporangium (Fig. 7). Then 
another upper primary wall-cell soon divides by a similar wall. These cells 
divide further several times, thus contributing to the greater part of the 
Va 
Text-figure. Longitudinal section of a 
young sporocarp. in , indusium ; a, air chambers ; 
m.i ., microsporangium-initial. 
