156 MR A. ANSTRUTHER LAWSON ON 
an enlargement of the nuclear vacuole, that during this period there is an increase in 
the amount of contained karyolymph. In spore mother-cells this enlargement is very 
great indeed, and constitutes a conspicuous growth period (Lawson, 1911), while in 
somatic cells it is not so great. Now, coupling these facts with the building up of new 
chromatin substance with each nuclear generation, there is a suggestion that the 
condition of the chromatin has an influence on the varying osmotic relations. For 
now, again, it would seem that as soon as the chromatin assumes the compact form of 
the chromosomes there immediately follows a diminution in the volume of the 
karyolymph, and this continues until the nuclear sap is completely diffused into the 
cytoplasm. 
This investigation, however, is not intended to cover all the phenomena of mitosis; 
its only object has been to throw some light on the problem of the achromatic spindle 
and the factors responsible for its formation in vascular plants. 
The writer wishes to express his indebtedness to the Executive Committee of the 
Carnegie Trust for a grant to defray the cost of the plates illustrating this paper. 
SUMMARY. 
A study of the microspore mother-cells of Disporum, Gladiolus, Yucca, Hedera, 
and the vegetative cells in the root tip of Alliwm has revealed a series of stages in the 
development of the mitotic spindle which have never before been described. 
The new stages that have been discovered are to be found in the prophase immedi- 
ately preceding the organisation of the equatorial plate, and concern the fate of the 
nuclear membrane. 
The interpretation of these stages has thrown a new light on the process of mitosis, 
and necessitates a revision of the accepted views of nuclear phenomena. 
Contrary to the generally accepted view, it has been found that the nuclear mem- 
brane does not break down or collapse at any period during spindle development, but 
behaves as one would expect a permeable plasmatic membrane to behave under varying 
osmotic relations. 
The nucleus is regarded as an osmotic system, and its membrane constitutes an 
essential element in that system. 
It is a fact of common knowledge that the chromatin changes both in quantity and 
form sometime before the metaphase. The chromatin must increase in quantity, because 
the same amount is present for each mitosis. It changes in form from the finely 
divided condition represented in the reticulum and spireme to the more compact and 
homogeneous form of the chromosomes. 
It would seem that these changes are in some way responsible for a variation in the 
osmotic relations of the karyolymph: at any rate, a gradual diffusion of the nuclear 
sap immediately follows these changes in the form of the chromatin. 
A series of stages has been found showing beyond doubt that, closely following the 
organisation of the bivalent chromosomes, there is a gradual diminution in the volume 
