726 RALPH S. LILLIE 
tively to the regions toward which the radiations converge; the 
nuclear region is positive also. The lines of force in closest prox- 
imity to the nucleus are thus curved sharply toward the latter 
and in their aggregate show a spindle-form. Others on the 
‘nuclear side show less departure from the straight radiating course 
of the great majority; a certain number, intermediate in position 
are less strongly developed and show a less well defined course, 
peculiarities probably due to their being under the influence of 
both fields which partly counteract each other’s action. 
As to the origin of these two fields,—they arise, on the present 
hypothesis, in consequence of simultaneous and similar changes 
in the electrical polarization of the two boundary surfaces of the 
cytoplasm, due to alterations in the ionic permeability of the 
membranes. Let us assume the primary change to be a rapid 
and decided increase in the ionic permeability of the plasma- 
membrane.’® This increase of permeability is assumed to be 
not uniform and simultaneous over the entire surface of the cell 
but to be most marked and most rapid over two extensive areas 
(e.g., between A and B, C and D, fig. 3) at the opposite sides of 
the cell adjoining the polar axis. It seems necessary to assume 
this definite localization of the areas of markedly increased perme- 
ability because of the characteristic and symmetrical bipolarity 
of the mitotic figure. If membranes are concerned in the process, 
a corresponding symmetrical and bipolar alteration of these struc- 
tures must be assumed. There is at present no definite and 
independent evidence that this is the case; but the assumption 
seems justifiable since it involves merely the extension of the 
50 The question as to whether under normal conditions the nuclear or the plasma 
membrane is the first to undergo increase of permeability 1s probably to be an- 
swered differently in the different cases. In the majority of dividing cells the radia- 
tions appear to originate near the surface of the nucleus, typically at points where 
the nuclear membrane begins to break down, i.e., where its permeability first unde- 
goes decided increase. This indicates that the primary change is frequently, 
perhaps usually, at the nuclear membrane. On the other hand, in artificial par- 
thenogenesis or normal fertilization the primary action is on the plasma-membrane. 
The essential principle is that alteration of the potential-difference at the one 
membrane involves a similar alteration at the other. The case is analogous to 
the stimulation of one muscle or nerve by the action-current of another—stim- 
ulation signifying increase in the permeability of the limiting membranes. 
