643 
Ovule of Bowenia spedabilis. 
Cycads a lysigenous cavity forms by progressive disintegration of cells and 
eventually becomes large enough for the storage and growth of the pollen. 
A very different and more elaborate type of development of the storage 
cavity or plinth has been assumed to take place in the Lagenostomales. 
The whole pollination of this group is in striking contrast to the simple 
development and arrangement of the pollen-chamber of Cycads, in which 
there is no suggestion at any stage of the previous existence of such 
a specialized type of nucellus as was present in the Lagenostomales. 
A closer comparison can, however, be made with the pollen-chamber 
of the seeds of the Medulloseae. 1 In Trigonocarpus , Stephanospermum , &c., 
the pollen-chamber consisted of a prominent nucellar beak which engaged 
with the base of the micropyle, the cavity of which led below to a larger 
chamber, where the pollen-grains developed. The seeds of the Cordaiteae 
also had a pollen-chamber of this type. Some of Brongniart’s 2 figures of 
apparently young ovules of Cordaiteae show the method of development 
of the pollen-chamber. In his Fig. 13, PI. Ill, he shows the apex of the 
nucellus of Cardiocarpus augustodensis , in which a median cavity, evidently 
formed by disintegration of tissue, leads to a larger cavity containing pollen- 
grains. The upper passage is described as a ‘ canal conduisant a la chambre 
pollinique’. From the appearance of the tissues in these figures it seems 
probable that the development of the pollen-chamber was upon the same 
lines as has been described in Cycads. Although there are no figures of 
young stages of Trigonocarpus it is probable that the pollen-chamber, 
which so closely resembled that of Cardiocarpus , developed along similar 
lines, and therefore along similar lines to the Cycads. 
As regards structure and development of the pollen-chamber, therefore, 
the closest parallel to the Cycads among fossil seeds is to be found in seeds 
of the Trigonocarpus affinity, and there seems to be no evidence in either of 
them that this structure had resulted from simplification of a more compli- 
cated type, such as that of the Lagenostomales. 
The general morphology of the Cycad ovule has long been a problem 
of interest, and since the discovery of the relationships of Lagenostoma, 
comparisons have been made between the two, which scarcely seem justified. 
The Cycad integument has been described as a morphologically double 
structure, 3 formed by the fusion of a cupule-like structure with a single 
integument, the line of fusion being in or near the stone layer. If such were 
the case it might reasonably be expected, that the development of the 
Cycad integument would show something indicative of its double nature. 
The integument of Bowenia develops as a single homogeneous tissue, which 
later gradually differentiates into the three characteristic layers, but shows 
no indication that it is morphologically a double structure. Similar 
1 Scott : Studies in Fossil Botany. Part II, 1909. 2 Brongniart, loc. cit. 
3 Stopes, M.C. : The Double Nature of the Cycadean Integument. Annals of Botany, vol. xix, 1905, 
U U 2 
