376 PROF. F. W. OLIVER ON THE STRUCTURE 
THE STRUCTURE OF THE POLLEN. 
This part of the subject has been reserved for special eonsideration, and the observa- 
tions drawn from the two species of seeds are given consecutively. 
(1) Pollen of Stephanospermum akenioides. 
In almost every pollen-chamber of this seed the ebaraeteristie pollen, so often 
mentioned by Renault, is present. These grains are usually congregated together in the 
apex of the chamber, as in Pl. 41. figs. 2, 3, 6, and Pl.42.fig. 15. It is only exceptionally 
that a grain has been drawn down to the level of the diaphragm (d.) (as in fig. 3, p.!) or 
below it, as with the isolated abortive grain (p.?) in the same photograph. The majority 
of the grains are large, possessing average dimensions of 160 ux 100 u, whilst occasional 
grains reach a length of 200 u or even 210 u (figs. 6 & 15). With these larger grains 
others of the same type are present, having an average length of 100,4. "These may be 
regarded as grains which have failed to develop after reaching the chamber. In buta 
solitary instance has foreign pollen been found in the chamber. This case is illustrated 
in fig. 15, where, in addition to the large normal pollen-grain, two small grains are 
seen lying in the apex of the chamber just above it, whilst a third overlies its 
left-hand extremity. The upper left-hand grain shows a longitudinal furrow and is 
doubtless **foreign "; so, too, is its right-hand neighbour, but referable to the Cordaitean 
type; the third little grain is of the same nature. — All these foreign pollen-grains seem. 
to agree in remaining abortive, contrasting in this respect with the true pollen of 
Stephanospermum, which has undergone further development is this pollen-chamber. 
The normal pollen shows an external finely granulated or sculptured coat, the 
exospore, whilst the contents are divided into a tissue of many cells. In this point 
of the multicellular nature of the pollen, concerning which there has been some 
question, I find myself in complete agreement with Renault, who described the grains in 
his * Cours de bot. foss.’ vol.iv. p. 185, and has adhered to his interpretation consistently 
ever since. The internal cells line the exospore everywhere, and surface views of the 
grains show the whole area mapped out by this internal tissue as in Renault’s figures *, 
and here in Pl. 42. fig. 15 and in fig. 14, p. 
A study of sections of the grains confirms the conclusions obtained from surface views. 
Pl. 42. figs. 14, p-, and 16, p.?, represent a section of the same pollen-grain cut at right 
angles to its major axis. Here we are looking into the grain without the intervention of 
the surface, ez. (fig. 16) being the cut edge of the exospore. Within the grain, which is 
unusually well-preserved (fig. 16, p.2), a system of three convergent radial walls is evident. 
The internal organisation of this grain is further elucidated in Pl. 43. figs. 30 and 31. 
Fig. 30 is a view of the cut surface magnified about 180 diameters. The cut edge of the 
exospore (ev.) is seen as in fig. 16, and the three radial walls a, b, and c; traces of 
two more, d and e, are also discernible. Fig. 31 is an optieal section of the same grain 
vu a deeper plane and shows the walls a, b, c, and d as complete membranes, whilst e 
T Notice sur les travaux scientifiques de M, B, Renault, 1896, pl. 6. fig. 8; Cours de bot. foss. iv. pl. 21. fig. 9. 
——— À—— — — a i 
