132 BOTANICAL GAZETTE [SEPTEMBER 
across the nuclear cavity. Other dense material within the nucleus 
is in every case deposited in a crescent-shaped mass against the 
nuclear membrane toward the positive electrode. This is shown 
in some of the cells of figs. 3 and 5. The chromatin appears very 
coarsely granular. Occasionally a cell is found in which a small 
amount of this granular material is deposited against the side of 
the nuclear membrane toward the negative electrode, leaving a 
clear central space across which lies the much flattened nucleolus. 
The cells of the central cylinder about 1 mm. from the cap 
show shrinkage which is evident at the ends of the cells, but not 
laterally. 
In the cortex of the first millimeter the cells show greater effect 
than in the central cylinder. The cytoplasm in these cells is in 
nearly every cell definitely aggregated against the wall toward the 
positive electrode. The nucleolus no longer lies across the nuclear 
cavity, but has migrated with the chromatin toward the positive 
electrode (fig. 8). The nuclear cavity is no longer spherical but . 
egg-shaped, with the smaller end toward the anode. Chromatin 
and nucleolus are packed into the small end, and seem to have 
forced distention of the nuclear cavity. Frequently fine granular 
threads radiate from this mass toward various points in the periph- 
ery of the nucleus, as shown in figs. 8 and 11. This bears a strik- 
ing resemblance to fig. 12, pl. 18, of Morrier’s paper (16). The 
nucleolus in most cells at this stage cannot be distinguished as a 
separate body from the chromatin. _ 
It is in dividing cells that the greatest effect of the current might 
perhaps be expected, yet such is not the case. Migration does not 
take place at all in a cell in the process of nuclear division in either 
cytoplasm or chromosomes. Staining of division figures is poor, 
and in most cases presents a blurred picture in which the chromatin 
has the appearance of having melted together. Cells with nuclear 
division are shown in figs. 2 and 3. This blurred condition is 
characteristic of all division figures, no matter at what stage. All 
parts of such a cell usually stain a deep red with saffranin. The 
absence of migration should be expected in such cells from results 
found by Kite (13) and CHampers (4), who both found the pro- 
toplast during division in a very viscous state, in the form of a gel. 
