624 SUMMARY OF CURRENT RESEARCHES RELATING TO 
those of Strasburger. At an early stage the parietal protoplasmic layer 
of the embryo-sac contains a single layer of lenticular nuclei which are 
connected together by a number of radiating strands of protoplasm. 
Between the nuclei, and in the midst of the radiating strands, are the 
cell-plates, i. e. plates composed of isolated granules of protoplasm. 
When the formation of endosperm begins these cell-plates are trans- 
formed directly into the cellulose-membranes which eventually divide 
the parietal layer of nuclei into a number of polygonal cells. The 
process is somewhat intermediate between that of ordinary cell-division 
and that known as free-cell-formation. While this process is taking 
place the embryo-sac of Gymnosperms is a free cell with rounded 
outline, sufficiently large to be visible to the naked eye. By the 
gradual increase of the parietal layer, the embryo- sac eventually becomes 
completely filled up by a compact tissue of remarkable regularity, 
the cells of which are mostly elongated greatly in the radial direction. 
In smaller details four different types are described, represented by 
Cephalotaxus, the Cupressinese , Taxus , and Ephedra. 
The appearance of the cell-plates above described coincides with 
the fragmentation of the large and dense elements of the chromatin 
into fine granules. The authoress agrees with Frommann and Heuser, 
but differs in this respect from Strasburger, in asserting that the fila- 
ments of the nuclei in the parietal layer are intimately united with 
those of the surrounding protoplasm, forming, with them, a continuous 
net-work. The substance of which the cell-plates are composed consists 
of granules, the denser and larger of which are coloured by methyl- 
green with the same intensity as the nucleoles and other denser elements 
of the nucleus. The number of primary cells of the endosperm usually 
corresponds to that of the original parietal nuclei of the embryo-sac, 
and it appears certain that these are the essential factors in the cell- 
formation which takes place within the sac. 
It is a group of short cells belonging to the parietal layer which 
ultimately develope into the corpuscles or secondary embryo-sacs ; and, 
in the tendency, in Pinus and Cephalotaxus , towards the early differentia- 
tion of these cells, the authoress sees the foreshadowing of the process 
which is universal in Angiosperms, the formation of the embryonic vesicles 
before that of the endosperm. Ephedra exhibits a still closer approxi- 
mation in this respect to Angiosperms. 
Relations between Insects and Flowers.* — Mr. T. Meehan epi- 
tomizes his arguments in favour of the view that too much stress is laid 
by botanists of the Darwinian school on the part played by insects in the 
pollination of flowers. He affirms that flowers do not abhor “own- 
pollen,” and that no flowers are so truly fertile as the cleistogamous, 
while those which fertilize before the corolla expands are also certainly 
fertile. Plants wholly dependent on insects for fertilization are all 
perennials ; and an innumerable number of the flowers of these plants 
fall unfertilized. All annuals, on the other hand, though in some cases 
so arranged that cross-fertilization may occur, can self-fertilize when 
cross-fertilization fails; and in almost all cases annuals have every 
flower fertile. 
Bot. Gazette, xvi. (1891) pp. 179-7. 
