M. Koidzumi 
283 
characteristic flagellar cords arise. No special structure such as a basal 
granule is recognizable at the point of connexion of the cords and the filament. 
In the majority of cases the pointed extremity of the filament is found at the 
anterior tip of the body, and the cords arise directly from it at that point; 
but not unfrequently the pointed end of the filament lies a little within the 
endoplasm, and is then provided with a rather thick stalk connecting it to 
the anterior tip of the body, the cords arising from the end of the stalk (PI. XIV, 
fig. 57). The number of the cords is four in the young forms and eight in 
the mature ones. They run backwards, attached to the surface of the body, 
to the posterior end, where they become free. Porter remarked that the 
cords of the American species “can be traced to the posterior extremity of 
the animal, and thence back again on the opposite side of the animal to the 
anterior end.” This is not true of our forms. The free ends of the cords, on 
the contrary, are fairly long and distinct, and never escape the eye of a careful 
observer. The cords run in a left-handed spiral round the body. They are 
rather straight in small forms, while they are remarkably wavy in large forms, 
and the larger the body, the more conspicuous is their winding. The surface 
of the body is not even, but each cord is situated on a ridge. In small forms 
the ridges are rather indistinct; but they are distinct in large forms, especially 
at the anterior end, where they frequently appear as rather flat and elevated 
membranes resembling the undulating membrane of Trypanosoma. In the 
living animal the cords and their membranes display a continuous and 
beautiful rippling motion, the ripples passing from before backwards. 
In the majority of forms, the body is naked and destitute of any kind of 
appendages, but rarely large individuals invested rather thickly with fila¬ 
mentous bodies are met with. In the American species studied by Leidy 
some are naked, while others are thickly u ciliated,” and those studied by 
Porter seem to be invested thickly without exception. It seems certain that 
the filamentous appendages seen in our forms are not real organs of the 
animals themselves. 
Individuals lately detached from the gut wall are frequently provided 
with a knob-like structure, stained quite as deeply as the axial filament, at 
the anterior end of the body, where the pointed end of the filament is fixed. 
The knob is usually situated at the tip of the axial filament, but sometimes 
a short peduncle connecting them is observable. In some forms the knob 
assumes the shape of a sphere or an ellipsoid, and is unbroken in its outline 
(PI. XIV, fig. 54); but in others it is not smooth but prickly at its distal end 
(PI. XIV, figs. 57 and 62). Long and distinct splinter-like filamentous structures 
are also frequently visible at this end (Fig. 56). This split or frayed-out struc¬ 
ture is variable in size and shape; it is commonly single, but sometimes two 
or more are found. As regards the function of the knob, I am of the opinion 
that it serves, when the animalcule is attached to the layer of chitinous 
substance investing the inner surface of the intestine, to preserve a firm 
connexion. 
