AND AFFINITIES OF STEPHANOSPERMUM. 393 
other Lagenostomas of Williamson *, have a peculiar organisation that differentiates 
them from Stephanospermum and its allies, though they resemble the latter in their 
radial symmetry. Perhaps the most important feature of difference is the vascular 
system of the seed, which is zucellar in Stephanospermum but wholly integumental in 
Lagenostoma. When regard is had to the Cycadean character of the seed Lagenostoma, 
on the one hand, and of the vegetative organs known as Lyginodendron, the plant 
to which it belonged, on the other, little doubt can remain as to the close relations that 
must have existed between this plant and the ancestors of existing Cycads. 
In addition to the Lagenostomas, possibly other seeds from the English Carboniferous 
rocks may be referred eventually to this group of seed-bearing Fern-like plants (the 
Lyginodendrez). The seed Gefopsis, Renault, also may possibly belong to the same 
cycle of affinity—in any case it presents close agreement with Lagenostoma in several 
ways t. 
The affinities of the Stephanospermum-group of seeds, however, are not yet known 
with certainty, though there are reasons for supposing that one of its members, the 
English Zrigonocarpon oliveforme, may have belonged to the genus Medullosa i. 
Further, there is Mr. Kidston's recent discovery of a seed, not wholly unlike a member 
of this group, in connection with the pinnules of Neuwropteris heterophylla, a fern-frond 
also referable to the Medullose:e A. 
Hence the position in respect of the radially organised seeds is as follows :—The 
Lagenostoma-group have been definitely referred to Lyginodendron, and take their place 
as the seeds of forms intermediate between true Ferns and Cycads. The rich series of 
seeds that conform to the Stephanospermum-type, on the other hand, are not yet so 
definitely referred, but it is a reasonable anticipation to suppose that they will eventually 
—through the Trigonocarpons—find a place in the Medulloseze. 
It may perhaps be added that remains of Aleihopéeris pinnules and of Myelozylon 
petioles are exceedingly abundant in the pebbles in which Stephanospermum occurs; 
but that, whilst the association of the last-named with the two former is very close and 
constant, no evidence of aetual eontinuity has been discovered during the course of the 
present investigation. 
The other French seeds described by Brongniart |, Cardiocarpus, Rhabdocarpus, and 
similar forms, are by common consent attributed to the Cordaitez. As the organisation 
of these ** Platysperms" is very uniform, and actual continuity with Cordaitean tissues has 
been established by Renault in certain instances €, we may regard the position of these 
* Williamson, * Organisation of the Fossil Plants of the Coal-Measures: Pt. VIIL," Phil Trans. vol 97, 
p. 233. 
t See a forthcoming memoir in conjunction with D. H. Scott, F.R.S., “On the Structure of Lagenostoma 
Lomari, &e.” 
+ G, Wild,“ On Trigonocarpon oliveforme,” Trans. Manchester Geol. Soc. vol. xxvi. 1900 ; D. H. Scott, ** On the 
Origin of Seed-bearing Plants,” Lecture delivered before Royal Institution of Great Britain, May 15, 1903, p. 11. 
§ E. Kidston, F.R.S., * On the Fructification of Neuropteris heterophylla, Brongn.,” Proc. Roy. Soc. vol. lxxii. 
(1904), p. 487. 
l| Loc. cit. © «Flore carbonifère, p. 311. 
SECOND SERIES.— BOTANY, VOL. VI. 3L 
