Nuclear Division in the Hepaticae . 503 
of which is separate from all the rest, and rather resembles 
a little heap of coarse sand when treated with reagents which 
specially stain the chromatic elements of the nucleus. Later 
on, however, as the granular aggregations become more pro- 
nounced, the connecting groundwork of linin is also visible. 
Each chromosome-rudiment subsequently assumes an ellip- 
tical shape, which eventually passes over into the well-known 
rod-like shape, with a hump or swelling occupying the place 
where the fission finally takes place. After the separation 
of the daughter-chromosomes, a cell-wall is formed across the 
spore-mother-cell, and the two daughter-nuclei become en- 
closed in proper nuclear walls before entering on the final 
mitosis. The nucleolus also reappears within each of them. 
Owing to the very small size of the objects in question it was 
not found possible to trace certainly the earlier stages, but 
at the time when the aster condition is reached, the two 
dividing nuclei precisely resemble the original one in the spore- 
mother-cell. Thus it would seem probable that two heterotype 
divisions succeed each other here, instead of a difference 
obtaining between the first and second mitoses, as in Lilies. 
After the completion of the process the spore-mother-cell is 
divided into the four special mother-cells, and the protoplasm 
within each one becomes surrounded with its own proper 
membrane. The spores are set free, as in other forms, by the 
disintegration of the original walls. It is very difficult to 
follow this in Lophocolea , but I have found either Bismarck 
Brown or Orange G. a good staining reagent for this purpose. 
Frullania dilatata was also examined. It exhibited a very 
distinct quadripolar spindle, but the cell-contents are so ex- 
ceedingly difficult to fix that I was unable to obtain any 
preparations of sufficient excellence to enable the entire 
process of karyokinesis being satisfactorily followed out. This 
is to be regretted, as the structure of the sporogonium in the 
Jubuleae section of liverworts is rather distinct from that of 
the others. But I saw nothing which would lead one to sup- 
pose that the process of spore-formation differs to any marked 
extent from that in the other Jungermanniae which I examined. 
