AND AFFINITIES OF STEPHANOSPERMUM. 377 
is broken. From a study of this and similar preparations it would appear that there 
were about five rows of wedge-shaped cells disposed symmetrically about the longer axis 
of the grain. Turning now to a longitudinal section of a grain, we may consider in 
detail that marked p. in Pl. 41. fig. 6. The details are fully illustrated in Pl. 44. 
figures 32, 33, and 34. This grain is not cut in actual longitudinal section, but it 
is very nearly so. Figs. 32 and 34 represent the grain as seen from above and below 
respectively. The exospore (shaded) and its cut edge are evident in both figures, as also 
are the remains of the internal system of cell-walls, laid open by the section. Fig. 33 
represents an optical section lying between the views rendered in figs. 32 and 34. Here 
are shown some eight cells lying around a central space (c) in which rests a dark 
nucleus-like body (x). The walls of the lining cells here show a copious perforation 
(as at w, fig. 33). 
Another longitudinal section is represented in fig. 35, a drawing of pollen-grain p.? 
(Pl. 42. figs. 14 & 16). As before, the cut edge of the exospore is shown (ex.) with traces 
of the internal cells (c, c), whilst in the middle of the figure is a spider-like body (cb), no 
doubt the same as 2 in fig. 33, but better preserved and also more typical—at any rate 
in the sections which have passed through my hands. The relation of the “legs” of the 
“spider” to the peripheral cells is suggestive, whilst at one place (^) the central body 
is continuous with the membrane. 
The interpretation which may be placed on all these appearances (7. e., those shown in 
figs. 30 to 35) is as follows :—The pollen-grain contains some five rows of internal cells 
(making some twenty cells in all), somewhat wedge-shaped in form, with the tapering 
apices of the wedges directed inwards. "These cells, prior to mineralisation (whether in 
life or death is uncertain), underwent a certain degree of separation from one another at 
the centre, a separation which involved a fracture of the walls in that region. In this 
way there was left behind a sort of replum, which has been compared to a nucleus in 
fig. 33, and to a spider in fig. 35. This replum is far from uncommon. ‘The transverse 
section (Pl. 42. fig. 16, p.?, and Pl. 43. figs. 80 & 31) is perhaps not cut quite across the 
equator of the grain, but may coincide with a plane a little nearer one pole, as indicated 
at d, fig. 35. Finally, the polien-grain p.!, in fig. 14, is cut in transverse section, 
the transverse zone corresponding to that which would be included between the planes 
e and f, fig. 35. This section, according to the focal plane examined, shows either a 
free spider-like body or a condition of affairs resembling figs. 30and 81. It thus removes 
all doubt as to the nature of the central body or “ replum," which consists of the coherent 
internal margins of the septa. 
Recently, M. Renault * has directed attention to the pollen-grains of Stephanospermum 
akenioides and ZEtheotesta. In both cases the grains are similarly organised, 7. e., many 
peripheral cells surrounding a central space, which often contains a nucleus-like body. 
In the case of Ætheotesta the grain is described. To quote the passage T :—“ Celles-ci " 
(i. e., the peripheral cells) ** paraissent groupées assez régulièrement autour d'une cellule 
* “Sur quelques pollens fossiles," Soc. d'Hist: nat. d'Autun, Séance du 13 juin, 1901, “Sur quelques Crypto- 
gamcs hétérosporées," ibid., Séance du 22 sept., 1901. : 
T From the paper of Sept. 22, 1901, p. 7. | 
SECOND SERIES.— BOTANY, VOL. VI. 3I 
