1904] LAND—EPHEDRA TRIFURCA 3 
ancestry. Atavistic tendencies should not be lightly passed over, for 
it is from such reversions that we may expect valuable hints as to 
previous conditions. 
THE MICROSPORANGIUM. 
In the first material collected, December 20, 1902, the group of 
cells which gives rise to the staminate flower is shown in jig. 2. No 
archesporial cells are yet distinguishable either by size or staining 
reactions, nor is the beginning of the perianth visible. Cell division 
is proceeding quite rapidly, and there is apparently considerable 
activity throughout the winter months. 
A month later the perianth is quite well developed (jig. 3). The 
cells immediately beneath the epidermis are about the same size as 
the adjacent ones, and as yet no differentiation into archesporium 
can be recognized. At the base of the strobili the flowers are much 
farther advanced than at the apical region. Later in the season all 
stages from rudimentary sporangia to mother-cells may be found in 
the same strobilus. . 
Fig. 4 shows a later stage in the development of an anther taken 
from the base of a strobilus. There is no positive evidence that the 
archesporium rises from a single hypodermal cell, but such is probably 
the case. The primary wall layer divides periclinally, giving rise 
to the wall layer and tapetum. The wall-cells do not divide peri- 
clinally, all divisions being anticlinal (figs. 5, 6). ‘The sporogenous 
cells do not divide in any definite plane (fig. 5). The stages shown 
in fig. 5 were common in basal sporangia February 9, 1903. 
Fig. 6 shows a more advanced stage of the sporangium. The 
wall layer and tapetum are completely separated, and the sporogenous 
‘cells have ceased to divide; in fact, they are ‘spore mother-cells which 
have not yet taken on the appearance which is characteristic of older 
mother-cells (February 15, 1903). A little later the wall-cells become 
flattened by the growth of the mother-cells and the tapetum. No 
further anticlinal division of the wall-cells takes place after they 
begin to be flattened. ‘They become much stretched by the further 
growth of the adjacent cells. The tapetal cells increase in size and 
stain intensely, this last because of the presence of food in large 
quantities. Fig. 7 shows a more advanced stage. The wall-cells are 
