236 C. CLIFFORD DOBELL. 
able degree of complexity. The nucleus itself may be regarded 
as consisting of three pairs of structures. These all lie at the 
anterior end of the animal, and are arranged roughly in the 
shape of a horse-shoe. At the extreme anterior end are two 
minute granules of chromatin, lying side by side, connected 
with one another by a delicate filament of chromatin, or else 
in close apposition. Immediately behind this pair, and 
united to it, is another pair of chromatin granules. These 
are also connected with one another across the middle line. 
It will thus be seen that the two pairs of granules form the 
four corners of a minute square area, free from chromatin 
(cf. fig. 29). The main part of the nucleus consists of a large 
lobe of chromatin on either side, connected with, and extend- 
ing backwards from, the posterior pair of chromatin granules. 
Extending backwards from the posterior pair of granules 
are two delicate rod-like structures, which I believe to be 
homologous with the axostyle of trichomonads. I shall there- 
fore employ the same name to describe them. Hach axostyle 
terminates at the extreme posterior end of the animal in a 
minute chromatic granule. The eight flagella arise as follows : 
From the anterior end six, from the posterior two. The 
anterior take origin from the two pairs of chromatin granules, 
one pair of flagella arising from the anterior, and a single 
flagellum arising from the posterior on either side (cf. fig. 29). 
The posterior flagella, or, as I shall call them, the caudal 
flagella, arise from the chromatin granules at the posterior - 
extremities of the axostyles. The length of the flagella is 
variable, but is frequently great. I have not unfrequently 
found individuals in which the caudal flagella had attained a 
length of over 30, or more than three times that of the body, 
In consequence of their length and the minute dimensions of 
the animal there is often great difficulty in counting these 
appendages, 
The axostyles are normally parallel, but they frequently get 
crossed, owing to the twisting movements of the animal (see 
fig. 30). ‘This crossed condition, therefore, cannot be regarded 
as the normal condition, though the Octomitus in the rat 
