766 PROFESSOR W. C. WILLIAMSON AND DR. D. H. SCOTT ON THE 
the vascular bundles in the petiole and rachis. This fact holds good for both genera. 
Both in structure and form, then, the leaves of Lyginodendron and Heterangiwm are 
Fern-leaves, their one point of difference being the occurrence of pitted as distinguished 
from scalariform tracheides in their xylem. 
Now, as regards Lyginodendron, the distinctively Fern-like character is limited to 
the foliage ; this, however, is not the case with the other genus. Apart from the 
secondary growth, the stem of Heterangium is essentially that of a monostelic Fern 
of the Gleichenia type. The resemblance of the stele of Heterangiwm to that of a 
Hleichenia is, in fact, very marked, as regards the distribution of the tracheides and 
conjunctive parenchyma, the position of the protoxylem-groups, and the presence of 
a wide pericycle. We think the comparison with Gleichenia may prove to be a fertile 
one, and hope to pursue it further on another occasion. It is not at all unlikely that 
the Gleichenia type of structure may be of great antiquity. We see, then, that if we 
left the secondary thickening out of account, there would be no serious reason 
against regarding Heterangium as a fairly typical monostelic Fern. We notice, 
however, certain peculiarities : 
1. The presence of pitted tracheides. 
2. The collateral structure of the bundles passing from the stele to the leaves. 
3. The differentiation of the peripheral part of the stele into more or less distinct 
bundles continuous with the leaf-traces. The last named character is more marked 
in H. tilv@oides than in H. Griew. 
The occurrence of secondary thickening in a Fern-like plant is not in itself very 
surprising. We know that it takes place in a perfectly typical way, though not to 
any great extent, in the stems of Botrychium and Helminthostachys at the present 
day. 
The question now arises, What is the relation between Lyginodendron and Hete- 
rangium? The latter, as we have seen, shows a great preponderance of Fern- 
like characters. This does not seem to be the case in Lyginodendron, in which the 
stem, apart from the leaves, suggests anything rather thana Fern. Yet the essential 
differences between the two genera are limited to their steles. In typical specimens 
of Lyginodendron the whole interior of the stele is occupied by pith, around which 
we find a ring of scattered bundles. The most remarkable point, however, is 
the exact agreement in structure between the bundles of Lyginodendron and the 
peripheral strands, which form part of the stele in Heterangium. We said, above, 
that a Heterangium might be regarded as a Lyginodendron, with the bundles con- 
nected together by medullary xylem. Conversely we might look on Lyginodendron 
as a Heterangiwm in which all the primary xylem except the peripheral strands has 
disappeared. It seems as if in Lyginodendron, which is probably the more advanced 
type, as it appears to be the later in geological origin, only that part of the primary 
vascular tissue is developed which is in direct connection with the leaves. All the 
