M. Koidzumi 
299 
and that the style in Microspironympha is of the same substance as, and most 
probably continuous with, the contents of the tubule and the layer sur¬ 
rounding the nucleus. Furthermore, I conjecture that these structures in 
Microspironympha and Holomastigotoides, together with the nuclear sac in 
Teratonympha and the corbule in Trichonympha , are homologous with the 
axostyle and the structure enveloping the nucleus in Joenia, Devescovina, 
Lophomonas, and related forms. 
As for the other two genera of the Trichonympha series which I have 
studied, viz. Spirotrichonympha and Holomastigotes, no dividing stages were 
observed, and our knowledge of these forms is too scanty for discussion. But 
I believe that these forms represent types similar to, but simpler than, that 
of Holomastigotoides . There is a mass of dense protoplasm connecting the 
anterior ends of the rows of basal granules with the nucleus; and I suppose 
that the latter gives rise to all parts of the mitotic figure. 
Let us now consider the structure of the members of the Pyrsonympha 
series. The organization of these forms appears quite simple, but it is, in 
reality, not less peculiar than that of the Trichonympha series. I distinguished 
many species in this group, and have classified them in two subgenera of 
a single genus, viz. Pyrsonympha and Dinenympha. All of these species are 
undoubtedly of the same type in essential points of organization, showing 
differences only in degree of development or differentiation of some organs. 
To make this clear, a brief recapitulation of the structure in Pyrsonympha 
grandis —the species showing the characteristic organization most clearly 
may be given here. First, there is an axial filament , running in the long axis 
of the organism, and ending anteriorly at the tip of the body; secondly, there 
are the flagellar cords which arise at the anterior tip of this filament and run 
backwards, attached on the body wall, to the posterior end of the organism; 
lastly, there is the nucleus, enveloped in the nuclear sac , which is kept in 
connexion with the axial filament and the body wall. Towards the anterior 
end, the outer wall of the axial filament stains more deeply than the internal 
part with iron-haematoxylin, although the boundary between the two parts 
is not distinct and no definite zones are visible. At the anterior extremity, 
a sharp differentiation into these two constituents is frequently observable. 
In some individuals, however, a small portion of the anterior end appears to 
consist of an axis of feebly stainable substance with deeply stainable cords 
of almost uniform thickness regularly arranged along its surface. Thus it 
seems certain that the axial filament consists of two components, which are 
distinctly differentiated only at the anterior end. In small individuals of 
Pyrsonympha grandis , the axial filament is commonly homogeneous throughout 
its entire length. In the other species of smaller size, and in the species of 
Dinenympha , the axial filament also commonly appears homogeneous through¬ 
out its entire length. In some individuals, however, differentiation into an outer 
wall, staining intensely with iron-haematoxylin, and a feebly staining centra 
part, may be observed on rare occasions. The most important feature in the 
