THE ORIGIN OF THE SEED-PLANTS 53 
ted his conviction that a large number of such fronds, including Newro- 
pteris, Alethopteris and other important genera, having never been 
found with Fern-fructifications, could not have been Ferns 1). This ne- 
gative evidence, however, carried little weight with palæobotanists ge- 
nerally. 
From another side, however, proof was soon forthcoming that some 
of the apparent Ferns of the Carboniferous were very different from 
any known plants of that affinity. Anatomical investigation, in which 
Williamson took the lead, showed that in his Lyginodendron (now Ly- 
ginopteris) and Heterangium, genera with the foliage of Sphenopteris, 
the structure was in certain respects like that of a Cycad rather than a 
Fern, though points of analogy with the latter group were also obser- 
ved. Williamson, in 1887, speaking of these plants, suggested that 
„Possibly they are the generalised ancestors of both Ferns and Cy- 
cads.” 
Similar combinations of characters were found in other groups; thus 
Medullosa, originally described as an extinct genus of Cycads, was 
shown by the independent investigations of WEBER and RENAULT to 
have borne the Fern-like fronds of Neuropteris and Alethopteris. The 
anatomy of Medullosa presents, in fact, an apparent union of Cycadean 
and Filicinean features, though on quite distinct lines from those of Ly- 
ginopteris or Heterangium. Another totally different genus, Protopitys, 
was placed by Solms-Laubach in 1893 among plants which are interme- 
diate in their characters between Filicineæ and Gymnosperms. 
Potonié in 1897 established the group Cycadofilices to include such 
intermediate or indeterminate forms, and his proposal was widely a- 
dopted. So far, however, there was no satisfactory evidence as to the re- 
productive organs of any of the Cycadofilices, and it remained therefo- 
re doubtful whether they really represented a distinct class, or only a 
peculiar race of Ferns. 
Then, in 1903, came the discovery of the seed of Lyginopteris oldha- 
mia, identified by F. W. OLIVER by means of the glands on the cupule 
of the seed Lagenostoma Lomaxi, agreeing exactly in structure with 
those long known on the vegetative parts of the Lyginopterrs. The iden- 
tification was confirmed by other points of agreement, and by analo- 
gous cases observed in other forms. Thus the fact was established that 
1) D. Srur, „Zur Morphologie und Systematik der Culm und Carbonfarne,” 
Sitzungsberichte Akad. Wiss. Wien. Bd. LXXXVIII, p. 638, 1883. 
