CELL DIVISION BY FURROWING IN MAGNOLIA 389 
merly studied by the writer. In Magnolia, however, the thickening 
takes place more nearly uniformly over the entire surface of the cell, 
especially where the wall is not in contact with the walls of other cells 
(fig. 11). The writer does not find as great a degree of uniformity in 
the thickness of the cell wall as the figures of Guignard (8) indicate. 
It is ordinarily about one twenty-fifth of the diameter of the mother 
cell during the heterotypic division, and in interkinesis it thickens 
slightly, so that it becomes about one sixteenth to one twentieth of the 
diameter (figs. 13 and 23). During interkinesis the mother cells are 
in many cases elongated in the direction of the axis of the heterotypic 
spindle. 
The Furrowing Process 
In Nicotiana (6) there is neither a cell plate nor a furrow formed 
between the first and second nuclear divisions. In Magnolia, however, 
the formation of both of these structures is initiated, but neither is 
brought to completion before the homoeotypic karyokinesis. The 
former completely disappears, whereas the latter is arrested (figs. 
6 and 8). Guignard's paper indicates that this furrow remains where 
it is arrested during the entire second division ; but I am of the opinion 
(fig. 9) that, occasionally at leagt, it may recede somewhat. The 
furrow appears in the plasma membrane at the equator of the hetero- 
typic spindle at a considerable interval of time after the stage of the 
ephemeral orange zone. Meanwhile the nuclei have enlarged and 
separated considerably (figs. 5 and 22), and the nucleoli have been 
developed. The nuclei remain in almost this same condition until 
they flatten in preparation for the next division. During this period 
of resting of the nucleus the furrow is being formed. It is obvious 
that there is some difficulty in arranging a series of stages in furrow 
formation on the basis of nuclear changes when there is no such series 
of collateral processes with which to compare it. One finds a great 
variety of furrows; some appear in section as minute mucronations, 
others are broad invaginations, some have sharp edges, others are 
rounded; some are shallow, others are deep. Any arrangement of 
these in a series must be merely hypothetical, but an attempt at seria- 
tion is not without value. It may be supposed that the furrow begins 
as a sharp, knife-like edge which deepens gradually and broadens at 
the base. The sharpness may be retained until the furrow reaches 
the maximum depth attained before the homoeotypic mitosis. There- 
