386 PROF. F. W. OLIVER ON THE STRUCTURE 
their general characters, it follows that a considerable latitude in pollen-structure may 
co-exist within a fairly close cycle of affinity. Hitherto, two types of pollen have been 
recognised, the Stephanospermum (akenioides)-Codonospermum-ty pe, filled with cells, 
and the Cordaitean type, in whieh there is an isolated prothallial or antheridial group *. 
It is premature to say whether the former type corresponds generally to the radially 
organised seeds of this period, the latter to the flattened. The problem does not appear 
to have been worked out in detail. The existence of a case like S. caryoides, where the 
pollen conforms rather to what I have referred to as the Cordaitean type, seems to 
indicate that seed- and pollen-structure do not run parallel in the two cases, and that 'the 
relations of the two broadly eontrasting groups of seeds may be more complex than 
seemed probable. i 
Actually, the pollen of S. caryoides differs considerably from that of Cordaites. ‘The 
primary internal complex of two cells (which is shown both in the matrix and in t 
pollen-chamber) would seem to mark a reduction from the Cordaitean type. Whether 
the secondary complexes, which later on appear within these primary cells in S. caryoides, 
represent merely a postponement in the appearance of elements present in Cordaites, 
even before escape from the pollen-sac, it seems hardly profitable to discuss in the 
present imperfect state of our knowledge. 
CHARACTERS DISTINGUISHING STEPHANOSPERMUM CARYOIDES, n. Sp. 
STEPHANOSPERMUM CARYOIDES, n. Sp. 
This supposed new seed resembles S. akenioides, Brongn., in being circular in transverse 
section and in the possession of an apical crown and micropylar beak, though in an 
abbreviated form. As in that species, the perimicropylar area is closely set with hairs. 
Another feature in which the seeds agree is the continuous tracheal mantle of the 
nucellus. 
S. caryoides is a large seed, approaching a globular form. Length 151 mm.; 
diameter 121 mm. 
Its pollen-chamber is large and broad, with well-developed shoulders. 
'The peculiarities of the pollen have been noted above. 
On FERTILISATION AND THE TRACHEAL MANTLE. 
The course in the foregoing seeds during life may be supposed to be somewhat as 
follows :—The pollen is caught at the apex of the ovule by a drop of mucilage which in 
any case must have been secreted here. ‘There is no evidence to show what was the source 
of this drop, whether it welled up from the nucellar apex or was secreted at the micropyle. 
As the pollen-chamber developed and assumed the character of an extensive excavation 
in the nucellus, the peripheral tissue became sharply delimited from the central deliquescent 
area and formed the “shoulders.” This flank or shoulder no doubt served the purpose 
of holding up the roof of the pollen-chamber, so that it should not collapse—a function 
* Cf. Renault, * Flore carbonifère, 1879, pl. 17. fig. 15. 
