990] DIVISION OF CELL AND NUCLEUS IN LIVERWORTS 395 
this to the persistence of the protoplasmic strands after the dis- 
appearance of the connecting fibers. 
In order to see these plates as formed by the thickening of 
the spindle fibers, it was necessary to cut the sections two or 
three # in thickness; for at this stage the daughter nuclei lie 
very close together, and the protoplasm about them is exceed- 
ingly dense. Staining was protracted two or three times the 
usual length of time, and the excess washed out with great 
care, 
For my study there was placed at my disposal an abundance - 
of excellent material of a species of Anthoceros collected and 
fixed by Professor Mottier near Naples in April 1898. On 
account of the large size of the cells and the greater clearness 
with which the finer details were brought out in the preparations, 
this material proved to be more favorable than that obtained 
ftom the Indiana species, in which many details were brought 
out with difficulty. 
For the behavior of the centrosome in Marchantia we have 
only the abstract from Mottier (’98). He says: “In the bs ot 
lative cells of Marchantia I cannot assert with absolute certainty 
that a definite central body or centrosome exists in all cases, but 
Tbelieve that such is the case. In some in which the kino- 
_ plasmic radiations are densely stained the dark center seems to 
be merely the point of union of the radiations, but if the stain 
Washed out so that the radiations are almost colorless, a 
Well defined and densely stained central body is generally to be 
Seen,” 
— Itis further stated that the centrospheres are attached to the 
Mclear membrane and are almost diametrically opposite by ue 
time the chromosomes are differentiated or even earlier. 
Fatmer’s statement that the centrospheres exert a pulling strain 
sen the nucleus is substantiated. Frequently the nutes: Ht) 
ne is drawn out by their influence into slender beaks. 5 
or the study of karyokinesis in these vegerative i -? % 
& and rapidly growing stalks of the archegoniophore o 
chantia was selected. These were fixed in the Flemming 
