466 MISS T. L. PRANKERD ON THE STRUCTURE OF 
layer of a single structure of multiple origin. The term has since been 
applied to the whole of the free part of the integument (16, р. 203) and is so 
used in this paper. 
It will be noted that the term pollen-chamber is used throughout merely 
descriptively of the space in which pollen is actually found. The term has 
been employed to denote the specialized apical portion of the nucellus, but 
now seems hardly appropriate, since this differs considerably from the 
pollen-chamber of recent seeds, and is the obvious homologue of structures in 
other fossil seeds (cf. (15) pp. 19 & 32) in which pollen has never been found. 
We therefore revert to Williamson's original designation of lagenostome for 
this organ. 
In accordance with current usage, the term plinth is employed to indicate 
the convex, free part ol the nucellus on which the lagenostome rests ; and the 
modern terms megaspore and nucellar membranes are substituted for the 
old “ perispermic ” and * nueular.? 
III. DETAILED Structure. 
А. [ntegument. 
It will be convenient to deal with the histology of the various regions of 
the integument under the following heads : (1.) the epidermis, (ii.) the fibrous 
layer, (iii.) the soft tissue, (iv.) the fluted lining of the canopy. 
Of these, the first two form the shell which surrounds the whole seed 
practically unchanged from base to apex, while, as explained, the fourth only 
occurs in the upper third of the seed. 
(1.) pidermis.—This consists of a single layer of closely fitting, prismatic 
cells, which seem to have been serially arranged. The long dimension lies at 
right angles to the surface of the seed, and is about 45 и, though it increased 
towards the apex, where the outer walls were usually convex, and often 
papillate (Pl. 28. fig. 9), to as much as 70u. In surface view the epidermal 
cells were approximately square with an average breadth of 20, though 
towards the apex they became somewhat broader tangentially to the 
mieropyle. As usually preserved the cells have thick transparent walls 
and very dark contents, which in favourable cases may be seen to be in 
continuity with one another by very fine processes (Pl. 99. fig. 3). It seems 
scarcely possible that this appearance is entirely due to mineralization, and 
that the actual protoplasts were not in continuity during the life of the seed, 
though it is probable that the processes of decay and fossilization have 
rendered this continuity more apparent (cf. Wieland (19), p. 135), in a 
manner comparable with modern laboratory methods employed for the same 
end—namely, by swelling the walls, thus rendering them more transparent 
