472 Yasui . — 0 /z the Life-history of Salvinia natans . 
eight chromosomes. The walls are formed between the four nuclei, and 
thus four microspores are established (PI. XLI, Figs. 29-32). 
With the entering of the synapsis stage of the mother-cells, the tapetal 
cells commence to disintegrate, and soon the partition walls between 
individual cells disappear, the degenerating nuclei being embedded in 
the common cytoplasm (Fig. 15). As has already been described by 
Juranyi (73) and Heinricher (’82), the fully-formed microspores separate 
from each other and lie free, side by side with the degenerating tapetal 
nuclei in the mass of the tapetal cytoplasm, which now completely fills up 
the inside of the sporangium (Figs. 33, 34). 
The mature microspores later move towards the periphery of the 
sporangium, and the centre of the latter is occupied by a mass of vacuolated 
cytoplasm (Pringsheim (’63), Prantl (79), Heinricher (’82)). No trace of 
disintegrating tapetal nuclei is found at this stage (Fig. 35). The mature 
microspore has a thin endospore and a thick exospore. It contains a rather 
large nucleus at or near the centre, and abundant starch-grains (Fig. 36). 
Male Prothallium. 
The microspore begins to germinate inside the sporangium in the early 
spring. The material kept in the laboratory was found to commence 
germination at the end of January, and to continue it until about the 
middle of May. 
The development and structure of the male prothallium agrees on the 
whole with the description of Belajeff (’98). Belajeff studied the prothallium 
in toto and used the plasmolysis method in differentiating the outline of 
individual cells. He does not seem to have observed any dividing figures 
in the developing prothallium, and his observations lack histological details. 
The microspore first divides into two cells by a wall tangential to 
the surface of the sporangium (Figs. 37, 38). The inner or lower of the 
two cells then gives rise to the two unequal cells — the lower small root- 
cell and the upper large prothallium-cell (Fig. 39). Early investigators 
such as Pringsheim (’63), Arcangeli (76), and Prantl (79), who made 
some errors in describing the structure of the male prothallium, failed to 
observe any root-cell. Campbell (’87), who has studied the spermatogenesis 
of Salvinia more recently, was also unable to find the root-cell. He states : 
‘ Die Spore theilt sich durch eine Querwand und die untere der gebildeten 
Zellen bleibt ungetheilt, die vegetative Zelle des Prothalliums darstellend.’ 
Belajeff (’98) was the first to demonstrate the presence of the root-cell, but 
he was not certain when it is formed. 
After the first division or sometimes just before, the exospore splits 
along three radiating lines on the surface of the microspore. The section 
of the germinating microspore at this stage is shown in Fig. 37. 
