135 
The Morphology and Physiology of the Nucleolus. 
none is observable in the smaller part. Macklin observed exactly 
the same condition in the cells of the embryo chick grown in 
tissue cultures, where the binucleate condition arose by the nucleus 
dividing arnitotically. He saw similar nuclei fuse and then divide 
mitotically. The elongated nucleus in the cell CP has probably 
arisen by fusion of two nuclei such as those shown in the cell BC. 
The binucleate condition appears to arise owing to the necessity 
for an increase in the surface of the nucleus in contact with the 
cytoplasm. Such an increase would facilitate the interchange of 
substances between nucleus and cytoplasm, and no doubt in such 
cells there is intense metabolic activity going on. 
At fig. 18v is shown another nucleus from a cell of the same 
blastula. In this nucleus, the nucleolus is apparently fragmenting, 
and here considerable difficulty arises as to whether any of these 
nucleolar fragments are extruded into the cytoplasm. In Plate V, 
in most of the figures, certain granules labelled NX are shown 
coloured red. The greatest difficulty in the present research has 
been to ascertain whether these granules represent extrusions from 
the nucleus or whether they are fused metaplastic bodies. The 
difficulty arises because these granules stain almost, if not in some 
cases quite, identically with certain of the yolk granules. It may 
be suggested that in some cases they represent fused metaplastic 
bodies, for generally they are much larger than individual yolk 
granules ; but my main reasons for suggesting that they are derived 
from the nucleolus are, firstly, that in some cases there appears to 
be continuity between parts of the granules inside and parts 
outside the nuclear membrane ; and, secondly, that the parts of 
such granules farthest removed from the nucleus usually stain 
slightly basophil, and such a change of staining reactions is 
characteristic of nucleolar extrusions which have been described 
by other observers. While, therefore, I feel that there is fairly 
good evidence for believing that nucleolar extrusion does take 
place during development, I do not consider that the problem can 
be definitely solved by the study of fixed material. The ultimate 
solution lies in observations on the living cells, and the large 
amount of yolk substance in Limnxa stagnalis renders it unsuit- 
able for observations intra vitam. Some other type is necessary 
for this purpose which can be grown in tissue cultures, and I am 
at present investigating the subject for future researches. 
A typical gastrula is formed in Limnsea by the usual method 
of invagination. At later stages the mesoderm cells are cut off 
from endoderm cells around the blastopore, and come to lie be- 
tween the ectoderm and endoderm. Considerable nucleolar 
activity takes place during their growth and development. They 
are distinguishable from the other cells of the ectoderm and 
endoderm at early stages, by the greater oxyphility of their 
cytoplasm. 
