368 
Trypanosoma lewisi 
saw any sign of conjugation, and moreover the broad forms go on with 
their development in a very different way, so we are inclined to believe 
that the thin forms are incipient broad ones. 
Fig. 7 shows again a form such as that seen above (Diagram III, 
Fig. 5) with two nuclei and the blepharoplast between the two, but 
these forms are uncommon and we mention them as a curiosity. 
Another one is figured in Fig. 8. The flagellum is absent (or not well 
stained), the blepharoplast seems to be fused with the nucleus, but 
probably it is a mere supposition. In point of fact, the blepharoplast 
never disappears during the whole process of development. We men¬ 
tion this form because the resemblance with Prowazek’s “ ookinete ” is 
so striking. 
The broad, club-shaped forms are transformed into the “ large oval 
forms” (Figs. 10—12), the flagellum shortening up and the undulating 
membrane disappearing. The blepharoplast is situated beside or in 
front of the nucleus. In this stage of development the flagellates 
generally pass into the hindgut, but further development may take 
place in the midgut. The large oval forms become gradually shorter 
and broader (Figs. 13—16), the flagellum is very short and often 
difficult to stain, so that it is only to be detected as a bright line 
arising from the blepharoplast (Fig. 15). Sometimes no trace of a 
flagellum is to be seen (Fig. 16) but we doubt if it has really dis¬ 
appeared and are inclined to believe that it stains only with difficulty 
at this stage. 
These forms pass into the round forms (Fig. 17) which may be 
found already in the midgut, but which are abundant only in the 
hindgut and the rectum. 
The nuclear structure of all these forms is not essentially different 
from the normal structure. Sometimes the central chromatic mass 
(karyosome) disappears; which is to be seen in the preparations 
stained after Giemsa and Heidenhain (Figs. 10, 12, and 15). The 
basal granule of the flagellum is often well marked (Figs. 2 and 9) 
but if the flagellum is not well stained the basal granule is invisible. 
Curious effects of agglomeration suggesting conjugation were some¬ 
times observed (Fig. 18). 
Fourth day. No flagellates are to be seen in the midgut. In the 
pyloric region of the hindgut, free and attached forms are to be observed. 
The latter are attached to the gut wall or to each other, forming 
rosettes; they are broad without apparent flagellum or undulating 
membrane. Apart from these forms the rosettes contain active club- 
shaped flagellates and round forms (cf. Diagram VIII). 
