STUDIES ON PAKASrnO PROTOZOA. 
467 
be classified upon uiorpliological grounds alone. Alexeieff 
(191']) gives five characters, which, taken together, should 
form, he considers, a reliable basis for classifying tlie 
species of Trichomonas. These deal with such points as 
the dimensions of the axostyle, the structure of the nucleus, 
the distribution of the extra-nuclear siderophilous granules, 
etc., and they must be studied, lie says, on the adult 
flagellate. Now these are characters that are subject to some 
fluctuation, particularly if the organism be a parasite of more 
than one kind of host ; moreover, it is by no means always 
easy to decide which is the typical adult form. In the case of 
Trichomastix, two of the said characters, i . e . the para- 
basal body and the condition of the undulating membrane, 
are necessarily absent, and this renders the species determin- 
ation on morphological grounds increasingly uncertain. 
What makes the difficulty still greater is that slightly 
stronger differentiation of iron-h«matoxylin preparations 
produces such difl^erence in the degree to which the 
“ species ” characters are exhibited. 
In the present case two forms of Tetratrichomastix 
occur (vide infra), but I hesitate to dub them species. As 
these two forms of the flagellate never occur side by side in 
the same preparation, I am strongly of the opinion that the 
degree of intensity of the staining must be taken into account. 
For the present I propose to consider them as forms of one 
flagellate — Tetratrichomastix parisii, n. sub-gen., n. sp. 
Diagnosis. 
Tetratrichomastix n. sub-gen. Flagellate with axostyle 
and five free flagella, four anteriorly, and one posteriorly 
directed. 
T. parisii n. sp. Axostyle slender and rather poorly 
developed. Nucleus oval or round, (1) compact, rich in 
chromatin blocks (form A, PI. 32, figs. 30 and 31), or (2) feebly 
siderophilous, with a few scattered chromatin granules and 
a well-defined nuclear membrane (form B, PI. 32, figs. 32 and 
