A STUDY IN CHROMOSOME REDUCTION. 615 
There is no doubt in my mind that these stages (figs. 25, 28, 29, and 30) represent 
normal changes taking place in the living cell—changes that precede the separation 
of the halves of bivalent chromosomes. It seems to me that the only explana- 
tion that will rationally account for the diminishing nuclear fluid is that based on 
osmotic diffusion—that the karyolymph has exosmosed into the cytoplasm by 
means of the plasmatic membrane,* and that the latter is functional in this way 
until the diffusion of the karyolymph is complete and the membrane in question 
envelops the chromosomes. Throughout these changes it should be noticed (figs. 25, 28, 
29, and 30) that as the nuclear vacuole becomes smaller and smaller the differentiated 
threads of cytoplasm become more numerous, longer, and more sharply defined. 
If we compare the conditions represented in figs. 29 and 30 with those represented 
in fig. 24, it becomes perfectly obvious that a change has taken place in the relative 
positions of the nuclear and cytoplasmic substances; that in this readjustment the 
cytoplasm comes to occupy a cubical space that is half as great again as that which it 
previously occupied. It seems very improbable that such great changes as these could 
take place without in some way affecting the configuration of the cytoplasm. It seems 
even more improbable that the large nuclear vacuole—as, for instance, that in fig. 24— 
should diminish and finally vanish without generating a state of tension in the cytoplasm. 
From a study of this and other types it was invariably found that the form of the 
cytoplasm is effected by these changes, and that a state of tension is set up by reason of 
the cytoplasm being obliged to occupy a greatly increased cubical space. This manifests 
itself by a marked change in the configuration of the cytoplasmic reticulum, which 
takes the form of drawn-out threads or fibrils—drawn out from the reticulum by the 
receding nuclear membrane. There is not much doubt in my mind that if the changes 
occurring in the living cell could be accurately followed, this explanation for the origin 
and formation of the spindle fibrils would prove to be the correct one. As it is, I 
consider it the most rational explanation that has yet been offered from interpretations 
of fixed material. I therefore find myself unable to accept the view so often expressed 
that the achromatic figure (or “ kinoplasm”) takes an active part in mitosis. I believe 
these fibrils merely express a state of tension which is caused in the first place by nuclear 
osmotic changes. 
If we now examine fig. 29, we find the karyolymph has been almost completely 
diffused into the cytoplasm, and as a result the nuclear membrane closes in about the 
bivalent chromosomes which are now crowded closely together. In the stage shown 
in fig. 30 there is still less karyolymph present, and the nuclear membrane is no longer 
* FARMER (1912). In his recent article on “ Nuclear Osmosis and its Assumed Relation to Nuclear Division,” Professor 
FARMER accuses me of using the word permeable instead of the word semi-permeable. He states: “The author continually 
calls it a permeable membrane.” Such a statement as this is not quite in accordance with the facts, and is very misleading. 
The term that I have continually used throughout the paper is a permeable plasmatic membrane. I have used this term 
advisedly, knowing full well that most cytologists believed that the plasmatic membranes of the cell were semi- 
permeable. I prefer to continue the use of the term plasmatic membrane in this connection and in the sense generally 
understood by cytologists. 
