British Coal Measures. 
93 
and a peculiarly ornamented testa. The fourth new type, R. elongatum 
(PL VII, Figs. 42, 43), is cylindrical in form, with faintly marked, rather 
distant ribs. 
I have also doubtfully referred to Radiospermum the Carpolithes mar - 
ginatus of Artis (PL VIII, Fig. 46), a very obscure and little known type, the 
testa of which is quite destitute of any characters which can be cited as 
distinctive. There may be room for difference of opinion as to whether 
this seed was radiospermic or platyspermic. The few specimens which 
I have seen are all compressed, but I think it probable that this is due to 
pressure, and that the seed was originally radiospermic. The narrow mar- 
ginal rim seen in some examples is probably also a pressure effect. 
N eurospermum , gen. nova. 
This type of seed, which is undoubtedly radiospermic, has in two cases 
been shown by Dr. Kidston 1 to belong to Medullosas with the Neuropteris 
type of foliage. The seed of Neuropteris heterophylla , which has not been 
named by Dr. Kidston, though it has been compared with certain species ol 
Rhabdocarpus , a platyspermic seed, should, I think, be determined, for it 
more often occurs in the detached state than in continuity with N. hetero- 
phylla. As it is obviously impossible to apply the name heterophylla to 
it, I therefore propose to give it the name N eurospermum Kidstoni (PI. VIII, 
Fig. 47) and to regard it as the type of a new genus, in which the testa of 
the oval or subcylindrical seed possesses close, prominent, parallel, longi- 
tudinal striae. 
The seed of Neuropteris obliqua 2 has similar characters, but this seed 
has not yet been recorded in Britain. 
Pterospermum , gen. nova. 
There is a single specimen (V. 1183) in the Geological Department of 
the British Museum (Nat. Hist.) of a most remarkable seed (PI. VIII, 
Figs. 51, 52), in an ironstone nodule from the Middle Coal Measures 
of Coseley, South Staffordshire, which is the only example known from 
this country. The ironstone nodule has been broken into three pieces, 
each corresponding to the spaces between the giant wings of this 
seed. The whole can thus be fitted together, and the general form of the 
seed studied. Although there is no cross-section of the whole nodule, 
I deduce that the nucellus was probably of very small transverse area, and 
triangular in form, each angle being produced into a large flap-like wing, of 
equal length with the rest of the seed. The nodule has broken naturally in 
planes parallel to each wing, the planes of least resistance, and I do not 
1 Kidston (’04). 
Kidston and Jongmans (’ll). 
