STUDY OF SPIRAEA 
spore mother-cells into tetrads ; then the 
microspores round up, and at the time of shed- 
‘they have disorganized the tapetal cells 
have absorbed their contents. Of the two 
between tapetum and epidermis, men- 
I ed above, the inner has now disappeared, 
g been absorbed by the tapetum for the 
“ft t of the sporogenous tissue. The outer or 
ecium has become a thick layer by rea- 
f the enlargement of its cells in which the 
teristic thickenings. 
¢ nuclei i in the pollen. grain (fg. « aI 
TRE MEGASPORANGIUM. 
€ comparative rate of development of sta- 
and ovules has already been stated. Long- — 
the appearance of the integume 
the tip of the ovule become enlar 2 
size and arrangement of conten 
cells (fig. 22). This mass_ of 
of Spiraea is comparable to th 
sburger.* Ensuing periclinal divi 
masses, the hypodermal being the t 
‘the sporogenous mass. Su 
ts from 
ogee length of the ovule | and ae | 
_are not conspicuous, but the walls have ; I 
At this stage’ there throu 
ightly developed on the placenta, = ° ae 
the tapetal layer by periclinal walls: 
he di 
457 
nding the spore mother-cells in a late synapsis stage, and 
surrounded by two layers of narrow, lightly staining wall 
This condition persists for some time as regards the 
raca Japonica 
] sectio 
‘spore ‘mother-cel x o 
ones 
nts, and while the ae : ie — = 
eral bypodermal 
those’ an neigh- oe 
heifer cells in the oe 
at found in Rosa livida 
ision tenulls in form- a 
ecessive ae 
