362 PROF. F. W. OLIVER ON THE STRUCTURE 
re-discovered the pollen-chamber, a structure at that time not generally recognised in 
Gymnosperms *. This observation led to others upon recent Cycads, whilst some years 
later we have the suggestion of Renault that the simpler Gymnosperms possessed 
antherozoids +, a forecast amply justified by subsequent researches upon Cycas, Ginkgo, 
and other genera. 
Perhaps no group of plants has ever travelled so rapidly to the forefront of botanical 
interest as have the Cycads. On the one hand, we have the great discovery of 1896 of 
the fact of their zoidiogamy, bringing the Gymnosperms into the closest relations 
with Pteridophytes. On the other, along the independent line of vegetative anatomy, 
Scott, co-operating in and continuing the investigations of Williamson into our Paleozoic 
fossil plants, has, through the Cycadofilices, shown the great probability of the Filicineous 
ancestry of the Cycads ł. This progress in our knowledge of the Cycads themselves, as 
well as of the fossil fern-like forms which reach up to them, renders a re-examination of 
the fossil seeds a pressing necessity. Though the general organisation of these seeds 
has been made known, there remains much in which uncertainty prevails. In particular 
may be cited the precise relation of the nucellus to its integument, the details of nucellar 
structure, the fate of the pollen in the pollen-chamber, and so on. There is, of course, 
the chance—a remote one, seemingly—that some lucky section may reveal evidence 
justifying the reference of a seed to the plant to which it belonged. 
Owing to the courtesy of Monsieur A. Roche, the Palzeontologist of Autun, it has been 
possible for me to commence the investigation outlined above. M. Roche has with great 
liberality put at my disposition a considerable number of uncut pebbles from Grand’ 
Croix as well as Autun. "These, together with others which I have collected in both 
localities, form the basis of the present communication. ‘Towards the cost of preparing 
microscopical sections of these fossils I am indebted to the Royal Society’s Government 
Grant Committee for a grant in aid. 
The present communication is restricted to a consideration of two seeds from the 
Stephanian of Grand’ Croix and to the bearing of the facts elicited. These seeds are 
Stephanospermum akenioides, Brongn., and an apparently new seed which it is proposed 
to name Stephanospermum caryoides. Brongniart, in his * Graines fossiles, divided the 
silicified seeds into two groups :—(1) Those exhibiting bilateral symmetry, more or less 
flattened, and probably of Cordaitean affinity; (2) Forms with radial symmetry and 
sometimes provided with longitudinal ridges, wings, &c. It is to the latter group that 
these seeds belong. 
In selecting Stephanospermum for first treatment I have been influenced partly by the 
faet that it was fairly abundant in the material at my disposal, but more particularly by 
the relative perfection of its preservation and the rather unusual characters of its 
organisation. In any case, a re-examination of Stephanospermum is appropriate on 
* The pollen-chamber with pollen in it appears to have been first figured by Griffith in 1854 (Te. Pl. As. tab. 378); 
see also Miquel, Journ. of Bot. vol. vii. 1869, p. 75. I am indebted to Sir William Thiselton-Dyer, K.C.M.G., for 
these references, 
t * Flore fossile d'Antun et d’Epinac,’ pt. 2, p. 277. 
t ‘Studies in Fossil Botany,’ Lecture xiv. 
