INTESTINAL TIUTANOPLASM AS OE FISHES. 
189 
slender forms move forward into the pi’oboscis-slieatb. They 
may occupy this situation as early as the sixth day after 
feeding. As time goes on they come to lie there in incredible 
numbers, and very generally attach themselves to the wall 
by their anterior end, i. e. the end with the free flagellum.’^ 
It is obvious that in the case of these preparations we are 
dealing with a specialised type of division resulting in tlie 
production of a more elongated trypanoplasma than the 
normal blood type (Miss RobertsoAs so-called comma” 
type), which is destined to develop into the elongate pro- 
boscis form, the true infective form. 
The form of Trypanoplasma which is apparently undergoing 
division in these preparations is shown in PI. 10, fig. 22, and 
the elongated infective stage in PI. 10, fig. 23. 
Unfortunately, in the preparations I have examined the 
early stages of division are very rare, though the later stages 
are exceedingly abundant. As far as I have been able to 
see, the new flagella arise by splitting, but this is at present 
a somewhat doubtful point. Whatever the origin of the new 
flagella may be, it is obvious, from PI. 10, figs. 24 and 25, that 
one blepharoplast after division travels backwards as the 
kinetonucleus elongates. The elongated kinetonucleus then 
divides by means of a simple transverse constriction. The 
division of the trophonucleus is precisely similar to that 
which has been previously described for Trypanoplasma 
congri, and it is interesting to note that here again, as is 
shown in PI. 10, fig. 29, the products of division may show 
for a short time the remains of the strand connecting the 
dividing trophonuclei. 
VII. Conclusions. 
It would appear that at present the nomenclature of the 
bimastigate flagellates is in a state of almost inextricable 
confusion. This state appears to me to depend largely on 
two factors — (1) the discovery of a free-living bimastigate 
form with a kinetonucleus (Prowazekia), and the consequent 
