218 Proceedings of the Royal Physical Society. 
sides facing one another (Fig. 4), and their outer sides are convex, forming — 
the basis of the polar caps. There are now two karyosomes present, 
presumably formed by the snapping across of the central axis, and they 
are embedded in the concave inner sides of the daughter plates. The 
daughter plates, no longer connected, draw apart and become inverted as 
regards their form. Their inner concave faces become convex and their 
outer convex faces become concave, still forming the bases of the pyramidal 
polar caps (compare Figs. 4 and 5). The daughter plates also rotate, form- 
ing an angle of about 45° with the longitudinal axis of the daughter cell 
(Fig. 5). Later the chromatin loses its thread-like appearance, the nucleus 
rounds itself off and is reconstructed. The karyosome is now in the 
centre of the granular chromatin of the nucleus, as shown in one of the 
nuclei in Fig. 6. At this point too the polar caps shorten and disappear 
finally. | 
Meanwhile the cytoplasm of Oxyrrhis marina has been dividing, as can be 
seen in the figures, until the form of a normal individual cell can be made 
out in each of the daughter cells. The division of the flagella was not 
followed, and no centrosome was seen. 
The Oxyrrhis marina in this culture were infected by a parasite 
allied to the Haplosporidia, and four of these are shown in Fig. 7, one 
dividing. ; 
The clearest figures of nuclear division in Oxyrrhis marina published 
hitherto were those of Keysselitz (4), drawn apparently from material left 
by Schaudinn. These figures are very small, do not indicate any polar cap, 
and show in the karyosomes very distinct centrioles, which I was unable to 
find in my preparations. ! 
Most of the accounts of nuclear division in the Dinoflagellata (of which 
Oxyrrhis marina is a primitive or degenerate member), show massive 
chromatin plates similar to those described above for Oxyrrhis marina, but 
the presence or absence of a karyosome, and its function if present, are points 
that appear to be doubtful in some of the cases. Borgert (1) figures some 
“ Nebenkérperchen” in the cytoplasm of Cezatvwm which, he thinks, probably 
divide independently of the nucleus. Jollos (3) decides that there is no 
centriole in Ceratium, and that the nuclear bodies described by him are 
plastin nucleoli, not centrosomes. Lauterborn (5) describes two to four 
nucleoli in Ceratvwm which on nuclear division taking place move to the 
poles of the spindle figure. A partial rotation of the daughter plates is 
. described, resembling that above described for Oxyrrhis marina. One of bis 
figures shows a faintly staining drawn-out body rather resembling the 
appearance of the karyosome in the daughter plate of Oxyrrhis marina. 
