214 BOTANICAL GAZETTE [MARCH 
a transverse row of superficial cells below the ligule.” Also, as regards 
the formation of the trabeculae and tapetum, Fitting’s account is identical 
with Smith's. 
The spore mother cell is distinguished by its finely granular protoplasm, 
large nucleus, and nucleolus. At one side of the nucleus lies a dense mass 
of coarse-grained protoplasm, in which are imbedded many small starch 
grains. Preceding the first division of the mother cell, radiations appear in 
the protoplasm, extending in all directions to the wall, but from no common 
center. The mass of mingled protoplasm and starch divides into two nearly 
equal parts, and new radiations appear between them as they separate. 
They finally take the positions of two foci of an ellipse, the spore mother 
cell being nearly of that shape. During this process the nucleus has shifted 
from the center to the periphery of the cell, and at its conclusion has returned 
to its original position. The two daughter masses (Zochter-Klumpen) 
elongate and lie in planes at right angles to each other and to the long axis 
of the cell. The starch grains arrange themselves in straight lines in each 
mass. Those near the middle slip toward either end and reunite in two 
groups, surrounded by the dense protoplasm. This process results in four 
masses of starch surrounded by the coarse granular protoplasm, arranged 
tetrahedrally. These changes the author followed in living material, observ- 
ing the spore through the sporangium wall and the several layers of sterile 
cells, tapetum, etc. The nucleus then divides by two rapid successive divi- 
sions, the spindle lying in such fashion that each of the four daughter nuclei 
lies by one of the Zochter-Kiumpen. New fibers arise from the surrounding 
protoplasm and extend themselves between the nuclei, thus forming a sex- 
tuple spindle. Equatorial cell-plates cross these spindles, cutting completely 
through the protoplasm in six planes from the center of the cell to its wall. 
Partition walls develop in these plates. It is evident that four of these 
walls have no connection with the spindles concerned in the division of the 
nucleus. 
The origin of the four megaspore membranes is worked out with great 
detail. The main points are as follows: While the four “* special mother” 
cells are still lying in the form of a tetrad enclosed by the mother cell mem- 
brane, each surrounds itself with a separate membrane called the «special 
mother cell”. membrane. The author did not determine whether this was 
formed by the mother cell membrane or by the protoplasm of the special 
mother cell. This thickens rapidly and divides into three lamellae which taken 
together constitute the erosfore. Between the exospore and the protoplasm 
of the cell (now called megaspore) developing from the latter, appears the 
mesospore. The outer layer of the exospore becomes roughened with spines 
or reticulations, and following their exact contour is laid down an incrusta- 
tion with much silica, which Fitting Styles the Perisfore. Some species of 
