M. Koidzumi 
277 
are laeotropic, it is hard to imagine that the one can grow into the other. 
As for the relationship between his ‘‘young form ’ and 'female form 
(Holomastigotoides), there are some points of resemblance in their organization. 
But they differ distinctly in the disposition of the basal granules of the flagella 
and the structure of the nucleus. No intermediate forms were recognizable, 
and the existence of a genetic connexion between the two is hardly con¬ 
ceivable. 
Although Spirotrichonympha leidyi occurs in all the individuals of Copto- 
termes formosanus which I have examined, it is not so abundant as some of 
the other species. Unfortunately I have been unable to find any dividing 
forms, or other stages in development. 
VII. Holomastigotes. 
This is a genus established by Grassi (1892), and described in his work 
with Sandias (1893), for a small organism found in Italian termites. He 
defined the genus more accurately later (Grassi, 1911) and published further 
details and figures in 1917. Forms closely resembling or probably identical 
with it are also found among the figures of the American trichonymphids, 
given by both Leidy (1881) and Porter (1897), but erroneously regarded by 
them as possible young stages of Trichonympha. An organism described and 
figured by Comes (1912, Fig. 1) as a “young Pyrsonympha with resting 
nucleus” also appears to belong to this genus. 
Closely similar forms also occur in the species of Leucotermes of Japan 
proper and of Formosa. The shape is variable in our organisms, and no sharp 
morphological distinction can be drawn between them and the I tab an or 
the American forms. As regards the finer details of structure, it is not possible 
to determine whether they show any differences or not, owing to the insuffi¬ 
ciency of previous descriptions. I propose, therefore, to call our form pro¬ 
visionally by the name of the Italian species described and figured by Grassi 
(1917) in his recent work. 
Holomastigotes elongatum Grassi (?) (Plate XIII, figs. 39 and 40.) 
This is a small and commonly fusiform or lanceolate form, measuring 
15-45 /x in length and 10-15 g in breadth. According to Grassi (1893) the type 
species attains a length of 70 g and a width of 24 g. The forms which I have 
studied thus appear to be considerably smaller, and should therefore, perhaps, 
be regarded as an independent variety. Grassi (1917), however, in his recent 
work does not state the dimensions of his species; but from his figures, of which 
the magnification is recorded, it appears that the organism may be con¬ 
siderably smaller than his original account would lead one to suppose. The 
body usually tapers towards the posterior end, and the anterior end usually 
projects slightly, forming a mammiform eminence. In this feature it appears 
to differ from the forms described by Grassi. The animalcule is, however, 
