NUCLEAR OSMOSIS AS A FACTOR IN MITOSIS. 139 
recorded in the process of spindle development as ordinarily accepted, and which [ 
have briefly outlined above. 
In the first place, the cytoplasm—presumably in the form of a reticulum—in the 
immediate vicinity of the nuclear membrane becomes transformed into a series of 
delicate threads or fibrils, which forms a dense weft about the nucleus. There seems 
not much doubt that these threads are really transformed cytoplasm, and this is the 
generally accepted view. It is believed that these kinoplasmic threads eventually 
become the fibrils of the mature spindle. But just how that transformation has been 
brought about and the factors responsible for the changes are questions which have 
never been satisfactorily answered. The kinoplasmic idea of Strasburger offers no 
explanation—it is simply descriptive. The hypothesis of electrical conditions and 
phenomena of induction as being factors in these changes is vague and difficult to 
comprehend. This idea, which has again been advocated by Harroe (1910), is no 
doubt very suggestive, but unfortunately it is based more upon appearances than upon 
knowledge of such electrical conditions in the living cell or upon the knowledge of the 
effect of such conditions upon living protoplasm. 
Another diffieulty—and in my opinion the greatest difficulty in the whole process of 
mitosis—that requires some explanation is the breaking down, collapse, and disappear- 
ance of the nuclear membrane. Many investigators (including the present writer) have 
either described or figured the breaking down of the nuclear membrane at a time when 
the multipolar stage has heen reached, or in somatic cells when the polar caps have been 
completely formed. This stage has been so frequently described that its actual occurrence 
has never been questioned. Now it is a fact of general acceptance that the nuclear 
membrane is cytoplasmic in nature (PEIRCE, 1903)—it is, in fact, the inner limiting 
layer of cytoplasm, and it is a membrane in virtue of its having come in contact with 
the karyolymph (Prerrer, 1890; Grkcorre, 1903; Lawson, 1903; YamManoucut, 
1906). Itis not only a plasmatic membrane, it is permeable and consequently osmotic. 
The nucleus itself is an osmotic system, and its membrane is the main essential element 
in that system. It should be remembered that at the time the membrane is supposed 
to break down, the nuclear cavity, while considerably under its maximum size, is 
nevertheless very large, and the membrane is under considerable tension, due to the 
high osmotic pressure of the karyolymph. Furthermore, it is well known (Lawson, 
1903; GATES, 1908; Farmer and Dicpy, 1910) that the membrane is capable of 
increasing or diminishing its surface area. ‘This is clearly demonstrated in any spore 
mother-cell where changes in the size of the nuclear cavity may be easily observed. So 
that, mm this sense, the membrane is both stretchable and elastic. The point that I am 
endeavouring to make is, that it is difficult to understand why a membrane with these 
properties should break down under the conditions of spindle formation. During the 
prophase the karyolymph is undoubtedly exerting a considerable pressure upon the 
membrane. ‘This is clearly indicated by the almost spherical form and turgid appear- 
ance of the nucleus at this time. One would expect a less violent process than breaking 
