642 Kershaw . — Structure and Development of the 
catching the pollen-grains. The breaking down of the nucellar tissue which 
would occur about this time, sets up a suction drawing the fluid and the 
pollen-grains contained in it down towards the nucellus. The grains are 
then accommodated in the small upper pollen-chamber, but they do not 
grow any further whilst in that position. By a further deliquescence of the 
cells of the central strand downwards and also outwards, a larger and wider 
cavity is formed — the lower pollen-chamber. The pollen-grains drop into 
this cavity and continue their development in it. There is sufficient room 
in this lower cavity for the stalk cells and generative cells to grow out from 
the pollen-grain and also a surrounding tissue of sufficient thickness for the 
rooting ends to burrow (Figs. 14 and 15, PI. LXI). The upper pollen- 
chamber, which is now functionless, appears as a pointed black cap of hard 
and shrivelled tissue on top of the lower pollen-chamber. 
The passage for the sperms to the archegonia is made partly by a further 
deliquescence of the cells below the lower chamber, and partly by the 
upward growth of the prothallus and consequent crushing of the nucellar 
tissue immediately above. In the abnormal ovules this upward growth of 
the prothallus and absorption of the nucellus is very marked ; it takes place 
to a less extent in the normal ovule, cf. Dioon . 1 The part of the nucellus 
marked out by the central strand in the young ovule, therefore, breaks down 
and deliquesces, and the remaining part, separating the sperms from the 
archegonia, is absorbed by the upward growth of the prothallus. 
Comparison with the Pollen-chamber and Ovule of 
Pteridosperms. 
Since the living Cycads are practically the only plants in which it is 
possible to study the actual development and working of the pollen-cham- 
ber, it is clearly necessary that the data should be critically examined with 
what has been described for that interesting region of the seed of Pterido- 
sperms. A comparison of the pollen-chamber mechanism of the Cycads 
with that of the Lagenostomales 2 emphasizes the relatively simple charac- 
ter of the former. Although two parts have been described above in the 
pollen-chamber of Cycads, which seem to correspond in position and function 
with the upper and lower parts of the pollen-chamber of these fossil seeds, 
they are not such differentiated regions as the c lagenostome ’ and ‘ plinth ’. 
There is no such specialization as one sees in the small cup-shaped lageno- 
stome of Conostoma with its sculptured walls, nor the circular cavity with the 
central cone of tissue in Lagenostoma. Again, the development of the 
pollen-chamber of the Cycads is a much simpler process than that which 
has been assumed to take place in Conostoma and Lagenostoma. In the 
1 Chamberlain, loc. cit. 
2 Oliver and Salisbury : Palaeozoic Seeds of the Conostoma Group. Annals of Botany, 
vol. xxv, 1911. 
