A STUDY IN CHROMOSOME REDUCTION. 617 
three phases * alone I consider the facts revealed sufficient to establish nuclear osmosis 
as a great and potent factor in mitotic phenomena. 
THe METAPHASE. 
As the nuclear membrane continues to recede, it becomes closely applied to the 
surface of the bivalent chromosomes, and each of these bodies is thus furnished with its 
own system of cytoplasmic fibrils. In another paper (Lawson, 19118) I have suggested 
how it is possible for this shifting of the lines of tension to result in the suspension of 
the chromosomes at the equator. I expressed it as my opinion that the osmotic 
surfaces, 2.e. the plasmatic membranes enveloping each chromosome, are determining 
factors in the suspension of these bodies between two sets of fibri] sheaves, and hence 
the formation of the equatorial plate. No new facts have been revealed in the present 
investigation to cause me to alter that opinion. 
There is no doubt in my mind that the fibrils composing the mature achromatic 
figure represented in fig. 31 have the same origin and structure as those of the earlier 
stages shown in figs. 30, 29, 28, and 25. If we now compare fig. 24 with fig. 31, the 
conclusion is irresistible, that the mature achromatic spindle of the metaphase is an 
expression of a state of tension due to the closing in of the large nuclear vacuole of the 
prophase shown in fig. 24. 
The main point of interest in the metaphase is the separation of univalent halves 
of the bivalent chromosomes and the movement of these halves (somatic chromosomes) 
in opposite directions. As shown in figs. 31 and 32, this separation is initiated by 
a lateral movement of the bivalent chromosomes, these structures spreading out in a 
plane parallel with the equator of the spindle. In fig. 33 a polar view of the equatorial 
plate is represented. From this and the preceding figures it will clearly be seen that 
the bivalent chromosomes are heteromorphic and take the form of double V’s. It will 
also be seen that the separation of the univalent halves (somatic chromosomes) 
commences not at the arms but at the bend of the V. A later stage in this separation 
is shown in fig. 34, where the chromosomes are represented on their way to the poles. 
The condition shown in fig. 36 is interesting, because it convincingly proves that the 
movement of the chromosomes is not to the poles themselves but in that general 
direction. It also shows quite clearly that the contraction of the fibrils plays no active 
role in this movement, for it may be clearly seen that many of the chromosomes pass 
beyond the region of the poles. This figure does not represent an isolated case ; many 
sections just like it were observed. 
In regard to this movement of the chromosomes to opposite sides of the cell, I can 
offer no positive explanation. From what we know of these chromatin bodies it would 
seem that there are times when they are attracted to one another; for instance, at the 
* Tn his recent review of my last paper Professor FARMER (1912) does not refer to either of these three important 
phases of mitosis. The discovery of the stages showing the persistence of the nuclear membrane throughout the 
prophase was also passed over without mention by the reviewer. 
TRANS. ROY. SOC. EDIN., VOL. XLVIII., PART III. (NO. 25). 90 
